Keywords: Signal pattern classification
Abstract: Low-frequency electroencephalographic (EEG) activity provides relevant information for decoding movement commands in healthy subjects and paralyzed patients. Brain-machine interfaces (BMI) exploiting these signals have been developed to provide closed-loop feedback and induce neuroplasticity. Several offline and online studies have already demonstrated that discriminable information related to movement can be decoded from low-frequency EEG activity. However, there is still not a well-established procedure to guarantee that this activity is optimally filtered from the background noise. This work compares different configurations of non-causal (i.e., offline) and causal (i.e., online) filters to classify movement-related cortical potentials (MRCP) with six healthy subjects during reaching movements. Our results reveal important differences in MRCP decoding accuracy dependent on the selected frequency band for both offline and online approaches. In summary, this paper underlines the importance of optimally choosing filter parameters, since their variable response has an impact on the classification of low EEG frequencies for BMI.

Keywords: Signal pattern classification, Data mining and processing - Pattern recognition, Data mining and processing in biosignals
Abstract: The Filter Bank Common Spatial Pattern (FBCSP) algorithm had been shown to be effective in performing multi-class Electroencephalogram (EEG) decoding of motor imagery using the one-versus-the-rest approach on the BCI Competition IV Dataset IIa. In this paper, we propose a method to improve the accuracy of decoding further through a rejection option based on the difference in the posterior probability computed by the Naïve Bayesian classifier. We applied the proposed approach on the BCI Competition IV Dataset IIa, and the results showed an increase in the session-to-session transfer accuracy from 0.824% to 14.0%, but the average decoded trials decreased from 93.2% to 34.2% using a rejection threshold between 0.1 to 0.9. We subsequently formulated a method to optimize the rejection threshold based on the maximum F0.5 score. The optimal rejection threshold yielded an average increase in accuracy of 5.1% with an average of 67.5% of trials decoded. The results showed the feasibility of improving decoding accuracy at a cost of rejection. Nevertheless, the results suggest that the use of reject option may be used as a training feedback system to train subjects’ overt and covert EEG control strategies.

Keywords: Signal pattern classification, Physiological systems modeling - Signal processing in physiological systems, Data mining and processing - Pattern recognition
Abstract: Brain computer interfaces (BCIs) offer individuals suffering from major disabilities an alternative method to interact with their environment. Sensorimotor rhythm (SMRs) based BCIs can successfully perform control tasks; however, the traditional SMR paradigms intuitively disconnect the control and real task, making them non-ideal for complex control scenarios. In this study, we design a new, intuitively connected motor imagery (MI) paradigm using hierarchical common spatial patterns (HCSP) and context information to effectively predict intended hand grasps from electroencephalogram (EEG) data. Experiments with 5 participants yielded an aggregate classification accuracy--intended grasp prediction probability--of 64.5% for 8 different hand gestures, more than 5 times the chance level.

Keywords: Signal pattern classification, Data mining and processing - Pattern recognition, Physiological systems modeling - Signals and systems
Abstract: Noninvasive brain computer interfaces (BCI), and more specifically Electroencephalography (EEG) based systems for intent detection need to compensate for the low signal to noise ratio of EEG signals. In many applications, the temporal dependency information from consecutive decisions and contextual data can be used to provide a prior probability for the upcoming decision. In this study we proposed two probabilistic graphical models (PGMs), using context information and previously observed EEG evidences to estimate a probability distribution over the decision space in graph based decision-making mechanism. In this approach, user moves a pointer to the desired vertex in the graph in which each vertex represents an action. To select a vertex, a “Select” command, or a proposed probabilistic Selection criterion (PSC) can be used to automatically detect the user intended vertex. Performance of different PGMs and Selection criteria combinations are compared over a keyboard based on a graph layout. Based on the simulation results, probabilistic Selection criterion along with the probabilistic graphical model provides the highest performance boost for individuals with pour calibration performance and achieving the same performance for individuals with high calibration performance.

Keywords: Signal pattern classification, Data mining and processing - Pattern recognition
Abstract: Surface electromyography pattern recognition methods have been widely applied to decode limb movement intentions for prosthesis control. These methods generally require amputees to provide sufficient myoelectric signals from their residual limb muscles. Previous studies have shown that amputees with high level amputation or neuromuscular disorder usually do not have sufficient residual limb muscles to provide enough myoelectric signals for accurate identification of limb movements. Electroencephalography (EEG), another bioelectric signal associated with limb movements has also been proposed and used for decoding the limb motion intents of humans. With an attempt to improve the performance of EEG-based method in identifying multiple classes of upper limb movement intents, four spectral domain features of EEG were proposed in this study. Motor imagery patterns associated with five different classes of imagined upper limb movements were distinctively decoded based on the four features extracted from 64-channel EEG recordings in four transhumeral amputees. Experimental results show that an average accuracy of 97.81% was achieved across all the subjects and limb movement classes. By applying a sequential channel selection method, an accuracy of around 95.00% was realized with about 20-channels of EEG. Thus, the proposed method might be potential for providing accurate control input for neuroprosthesis.

Keywords: Adaptive filtering
Abstract: Early detection of intention to move, at self-paced voluntary movements from the activities of neural current sources on the motor cortex, can be an effective approach to brain-computer interface (BCI) systems. Achieving high sensitivity and pre-movement negative latency are important issues for increasing the speed of BCI and other rehabilitation and neurofeedback systems used by disabled and stroke patients and helps enhance their movement abilities. Therefore, developing high-performance extractors or beamformers is a necessary task in this regard. In this paper, for the sake of improving the beamforming performance in well reconstruction of sources of readiness potential, related to hand movement, one kind of surface spatial filter (spherical spline derivative on electrode space) and the linearly constrained minimum variance (LCMV) beamformer are utilized jointly. Moreover, in order to achieve better results, the real head model of each subject was created, using individual head MRI, and was used in beamformer algorithm. Also, few optimizations were done on reconstructed source signal powers to help our template matching classifier with detection of movement onset for five healthy subjects. Our classification results show an average true positive rate (TPR) of 77.1% and 73.1%, false positive rate (FPR) of 28.96% and 28.74% and latency of -512.426±396.7ms and -360.29±252.16 ms from signals of current sources of motor cortex and sensor space respectively. It can be seen that the proposed method has reliable sensitivity and is faster in prediction of movement onset and more reliable to be used for online BCI in future.

Keywords: Ultrasound imaging - Elastography, Ultrasound imaging - Vascular imaging
Abstract: Rather than degree of stenosis, assessing plaque structure and composition is relevant to discerning risk for plaque rupture with downstream ischemic event. The structure and composition of carotid plaque has been assessed noninvasively using Acoustic Radiation Force Impulse (ARFI) ultrasound imaging. In particular, ARFI-derived peak displacement (PD) estimations have been demonstrated for discriminating soft (lipid rich necrotic core (LRNC) or intraplaque hemorrhage (IPH)) from stiff (collagen (COL) or calcium (CAL)) plaque features; however, PD did not differentiate LRNC from IPH or COL from CAL. The purpose of this study is to evaluate a new ARFI-based measurement, the variance of acceleration (VoA), for differentiating among soft and stiff plaque components. Both PD and VoA results were obtained in vivo for a human carotid plaque acquired in a previous study and matched to a histological standard analyzed by a pathologist. With VoA, plaque feature contrast was increased by an average of 60% in comparison to PD.

Keywords: Ultrasound imaging - Elastography, Ultrasound imaging - Other organs
Abstract: Acoustic radiation force impulse (ARFI) elastography is a non-invasive method for the assessment of liver by measuring liver stiffness. The aim of this study is to evaluate the accuracy of ARFI for the diagnosis of liver fibrosis and to assess impact of steatosis on liver fibrosis stiffness measurement, in rats model of non-alcoholic fatty liver disease (NAFLD). The rat models were conducted in 59 rats. The right liver lobe was processed and embedded in a fabricated gelatin solution. Liver mechanics were measured using shear wave velocity (SWV) induced by acoustic radiation force. In rats with NAFLD, the diagnostic performance of ARFI elastography in predicting severe fibrosis (F ≥ 3) and cirrhosis (F ≥ 4) had the areas under the receiver operating characteristic curves (AUROC) of 0.993 and 0.985. Among rats mean SWV values were significantly higher in rats with severe steatosis by histology compared to those mild or without steatosis for F0-F2 fibrosis stages (3.07 versus 2.51 m/s, P = 0.01). ARFI elastography is a promising method for staging hepatic fibrosis with NAFLD in rat models. The presence of severe steatosis is a significant factor for assessing the lower stage of fibrosis.

Keywords: Ultrasound imaging - Elastography, Image reconstruction and enhancement - Compressive sensing/sampling, Image retrieval
Abstract: Feasibility of applying compressive sensing (CS) to ultrasound radio-frequency (RF) data to produce elastography is investigated. The research also compares the performance of various CS frameworks associated with three common model bases (Fourier transform, discrete cosine transform (DCT), and wave atom (WA)) and two reconstruction algorithms (l_1 minimization and block sparse Bayesian learning (BSBL)) using the quality of B-mode images and elastograms from the RF data subsampled and reconstructed by each framework. Results suggest that CS reconstruction adopting BSBL algorithm with DCT model basis can yield the best results for all the measures tested, and the maximum data reduction rate for producing readily discernable elastograms is around 60%.

Keywords: Ultrasound imaging - Elastography
Abstract: This paper presents ultrasound elastography results obtained with a 640-element array transducer we have recently developed. This probe allows the acquisition of series of three adjacent imaging planes over time and therefore makes possible the computation of 2-D elastograms, with consideration of out-of-plane motion. In this study, elastography experiments were conducted on phantoms and bovine tissue samples, and compression was manually applied to the media via the hand-held ultrasound transducer. The results obtained with the proposed data acquisition and 3-D processing are presented and compared to those from a classical 2-D approach.

Keywords: Ultrasound imaging - Elastography, Ultrasound imaging - High-frequency technology
Abstract: Skin lesions change elastic properties near the surface. In the last decades, several non-invasive elastography techniques have been developed for detecting the mechanical properties of tissue. In particular, harmonic elastography is characterized for inducing shear wave propagation by an external vibrator in order to estimate shear modulus. However, near the boundary region, propagation is governed by surface acoustic waves (SAW). This paper combines crawling waves elastography with a high-frequency ultrasound (HFUS) system for the estimation of the SAW-to-shear compensation factor when ultrasound gel (US gel) is used as coupling interface. Experiments explore the SAW speed in a homogeneous phantom with a solid-water interface in order to corroborate theoretical findings. Subsequently, experiments in a solid-US gel interface are conducted in order to find the correct compensation factor. Preliminary results suggest that SAW propagation can be detected using HFUS, and shear velocity maps can be generated by applying the estimated empirical correction factor. This study will potentially avoid the underestimation of shear modulus when using SAW-based HFUS elastography which is promising for the better diagnosis of skin diseases.

Keywords: Ultrasound imaging - Elastography
Abstract: Mechanical properties of soft human tissue are linked to their pathological state. One way to assess these properties is through the Young modulus measurement, which is related to the shear wave speed in the medium. In order to characterize its elastic properties using sonoelastography, we introduce a new technique for shear wave estimation from a static interference pattern based on Shear Wave Holography. A relation between the mathematical representation of the interference pattern and the local shear speed is derived using the Phase Derivative approach. The experimental scheme is presented, detailing the advantages of the new configuration. Homogeneous and heterogeneous elastic media were simulated, generating an interference pattern on them. The shear speed estimation algorithm was explained and applied to obtain the speed map, calculating the mean value over each medium. The technique was tested on a homogeneous elastic phantom, yielding an estimation error of 6%. Overall, Shear Wave Holography using normal vibration is feasible and shows promising results in estimating shear wave speed in elastic materials.

Keywords: Magnetic resonance imaging - MR neuroimaging
Abstract: We used BrainVisa software in an exploratory analysis measuring the depth and sulcal profile of the central sulci of congenitally blind and sighted individuals. We found the greatest differences between the groups at locations on the central sulcus corresponding with the pli de passage fronto-parietal moyen (PPFM), suggesting a cortical reorganization of the primary sensorimotor area of the hand within the central sulcus. This may be in response to the congenitally blind individuals’ mastery of Braille or general increase of hand use in everyday life.

Keywords: Magnetic resonance imaging - MR neuroimaging, Image segmentation, Multimodal imaging
Abstract: Objective and non-invasive quantification of ischemic stroke and differentiation of salvageable from non-salvageable tissue is critical to treatment planning. However current Magnetic Resonance Imaging(MRI) techniques are time consuming and rely on manual detection methods. Computer aided preliminary screening of the injured tissue could assist neuroradiologists in performing more detailed analysis of the lesion components. An established Hierarchical Region Splitting (HRS) method was extended to segment lesions from adult patients who suffered a clinical stroke using diffusion- and perfusion weighted image (DWI-PWI) maps and associated computed maps. Apart from lesion quantification PWI-DWI based HRS was also able to automatically quantify core (irrecoverable) the penumbra (potentially recoverable) which helped to estimate salvageable tissue. The PWI-DWI/HRS results were validated by comparing with manually demarcated ground truth in terms of performance indices like lesion volume (82.1% accuracy), sensitivity (78.8%), specificity (99.3%) and similarity (78.54%) for a dataset of 10 acute adult stroke patients. Datasets were classified into severe, moderate and mild injuries based on total lesion volume. Proposed PWI-DWI/HRS method demonstrated accuracy close to manual lesion demarcation with high performance indices for core and penumbra in severe and moderate classes.

Keywords: Magnetic resonance imaging - MR neuroimaging, Brain image analysis, Functional image analysis
Abstract: Alcohol use disorders (AUD) are a major public health concern. Understanding the brain network alterations is of the utmost importance to diagnose and develop treatment strategies. Employing resting-state functional magnetic resonance imaging, we have performed a longitudinal study in a rat model of chronic excessive alcohol consumption, to identify functional alterations in brain networks triggered by alcohol drinking. Two time points were considered: 1) before alcohol consumption (control condition) and 2) after 30 days of alcohol drinking (alcohol condition). We first identified nine resting-state networks with group independent component analysis. Afterwards, dual regression was applied to obtain subject-specific time courses and spatial maps. L2-regularized partial correlation analysis between pairs of networks showed that functional connectivity (FC) between the retrosplenialvisual and striatal networks decreases due to alcohol consumption, whereas FC between the prefrontal-cingulate and striatal networks increases. Analysis of subject-specific spatial maps revealed FC decreases within networks after alcohol drinking, including the striatal, motor-parietal, prefrontalcingulate, retrosplenial-visual and left motor-parietal networks. Overall, our results unveil a generalized decrease in brain FC induced by alcohol drinking in genetically predisposed animals, even after a relatively short period of exposure (1 month). The only exception to this hypo-connectivity state is the functional association between the striatal and prefrontal-cingulate networks, which increases after drinking, supporting evidence in human alcoholics.

Keywords: Magnetic resonance imaging - MR neuroimaging, Image feature extraction, Brain image analysis
Abstract: Alzheimer’s disease (AD) is known as one of important diseases with world-wide impact. Unfortunately, we have not found the cause and treatment methods for AD. Magnetic resonance imaging (MRI) is an important research for AD research. Many recent studies focused on finding imaging biomarker associated with AD and mild cognitive impairment (MCI). Texture analysis jointly consider gray-level intensity and position of the voxels within a given region of interest (ROI). It can lead to better characterization of ROI than conventional approaches. We adopted one of the texture analysis methods called intensity size zone matrix (ISZM) to compare AD, late MCI (LMCI), early MCI (EMCI) and normal control (NC) subjects in both hemispheres of the hippocampus. As a result, we were able to distinguish among the sub-groups using the intensity variability feature and size zone variability feature computed from ISZM.

Keywords: Magnetic resonance imaging - MR neuroimaging, Brain image analysis, Magnetic resonance imaging - Diffusion tensor and diffusion spectrum imaging
Abstract: Conventional diffusion-weighted MR imaging techniques provide limited specificity in disentangling disease-related microstructural alterations involving changes in both axonal density and myelination. By simultaneously probing multiple diffusion regimens, multi-shell diffusion MRI is capable of increasing specificity to different tissue sub-compartments and hence separate different contributions to changes in diffusion-weighted signal attenuation. Advanced multi-shell diffusion models impose significant requirements on the amount of diffusion weighting (i.e. gradient coil performance) and angular resolution (i.e. in-scanner subject time), which commonly limits their applicability in a clinical setting. In this paper, we apply a high-b-value, high angular resolution multi-shell diffusion MRI protocol to a population of early multiple sclerosis (MS) patients and healthy controls. Through the Composite Hindered and Restricted Model of Diffusion (CHARMED) model, we extract indices for axonal density as well as parameters sensitive to myelin. We demonstrate increased sensitivity to microstructural changes in normal appearing white matter and in lesions in MS as compared to traditional models like DTI. These changes appear to be predominantly in axonal density, pointing towards the existence of axonal damage mechanisms in early MS.

Keywords: Magnetic resonance imaging - MR neuroimaging, Brain image analysis, Image feature extraction
Abstract: Statistical analysis using large-scale brain magnetic resonance (MR) image databases has examined that brain tissues have age-related morphological changes. The age of a subject can be estimated from the brain MR image by evaluating morphological changes with healthy aging. This paper proposes an age estimation method using local features of T1-weighted MR images. The brain local features are defined by volumes of brain tissues parcellated into 1,024 local regions defined by the automated anatomical labeling atlas. This paper also proposes the effective local feature selection method to improve the accuracy of age estimation. We evaluate the accuracy of the proposed method using 1,099 T1-weighted images from a Japanese MR image database. We also analyze effectiveness of each local region for age estimation and discuss its medical implication.

Keywords: Implantable systems, Implantable technologies
Abstract: Body area networks (BANs) are attracting great attention in healthcare and medical applications, and capsule endoscope is one of implant BAN applications. For real-time image/video transmission in implant BAN, we previously developed an in-body transceiver at around 30 MHz with a bandwidth of 50 MHz. However, an implant channel can cause severe waveform distortion so that the communication performance gets deteriorated. Therefore, in this study we developed an automatic equalizer on the transceiver. Moreover, we evaluated its communication performance. The measurement result shows that our developed equalizer can largely improve the communication performance in the implant BAN channel.

Keywords: Wearable antennas and in-body communications, Implantable technologies, Implantable systems
Abstract: Implant ultra-wideband (UWB) communication has so far attracted a lot of attention as a promising technology of reliable and high-speed transmission for implantable devices. In this paper, we developed a multiple-input multiple-output (MIMO) transmission system to realize reliable implant UWB communication. Furthermore, we carried out an experiment with liquid phantom for evaluating the developed UWB-MIMO system. Our experimental results showed that the develop system achieved the bit error rate (BER) performance of 0.01 at sufficient transmission points inside the liquid phantom ensuring the high date rate of 66 Mbps.

Keywords: Wearable antennas and in-body communications, Implantable systems, Physiological monitoring - Instrumentation
Abstract: One of the key issues about wireless technologies is their interaction with the human body. The so-called internet of things will comprise many devices that will transmit either around or through the human body. These devices must be tested either in their working medium, when possible, or in the most realistic one. For this purpose, tissue-like phantoms are the best alternative to carry out realistic analyses of the performance of body area networks. In addition, they are the conventional way to certify the compliance of commercial standards by these devices. However, the number of phantoms that work in large bandwidths is limited in literature. This work aims at presenting chemical solutions that will be useful to prepare a variety of wideband tissue phantoms. Besides, the colon was mimicked in two ways, the healthy tissue and the malignant one, taking into account studies that relate changes on the relative permittivity with cancer. They were designed on the basis of acetonitrile in aqueous solutions as described in a previous work. Thus, many scenarios could be developed such as multilayers which imitate parts of the heterogeneous body.

Keywords: New technologies and methodologies in medical robotics, Clinical robots
Abstract: In this paper, a new steering mechanism for wireless capsule devices is presented. The proposed system consists of a platform generating a magnetic field to direct and control the motion of a capsule. The platform contains an upper and a lower set of electromagnets. A permanent magnet is implanted inside the capsule to initiate the movement, which is set by the magnetic field delivered by the electromagnets. The total magnetic field at the capsule’s location is the sum of the contributions of each electromagnet. An experimental setup has been designed for testing and comparing between the performance of the capsule mobility in practice and simulations.

Keywords: Implantable systems, Wearable antennas and in-body communications, Wearable body sensor networks and telemetric systems
Abstract: This paper presents a swallowable sensor device that can be ingested orally, later arriving to the stomach, where the device can indwell for a long term. It can be egested at any time after it is triggered using wireless communication. This device is able to indwell by using a silicone balloon in the gastrointestinal tract. The balloon is inflated inside the stomach by a chemical reaction, and it is deflated to egest the sensor device using an actuator with electrolysis of water. Energy for the actuator with electrolysis can be fed wirelessly. A Near Field Communication and flexible antenna are employed for the power feeding and wireless data communication. The device size can be minimized without performance degradation because of the flexible balloon and the flexible antenna.

Keywords: Physical sensors and sensor systems - Optical and photonic Sensors and systems, Physiological monitoring - Modeling and analysis
Abstract: Stratum corneum is the outer most part of skin for barrier function. Adhesive tape stripping is a method to disrupt skin barrier function. In this study, we obtained NIR (Near-Infrared) spectrum of skin after tape stripping. Changes in skin spectra after barrier disruption were revealed by multivariate analysis of spectrum. NIR spectrum revealed that acute barrier disruption affected protein and lipid in stratum corneum

Keywords: Physiological monitoring - Instrumentation, Physiological monitoring - Modeling and analysis, Wearable sensor systems - User centered design and applications
Abstract: Using wearable, cuffless arterial blood pressure monitoring devices as examples, this talk will discuss the miniaturization, intelligence, networking, digitization, and standardization (MINDS) aspects of wearable medical devices. Specifically, some pulse transit time (PTT) based approaches including both machine learning methods and physiological modelling methods such as photoplethysmogram intensity ratio (PIR), multi-wavelength and multi-modal measurements will be examined for the improvement of cuffless blood pressure measurement accuracy. The focus of this talk will be placed on the studies of variabilities of blood pressure and its indicators. The application of physiological markers obtained from unobtrusive cuffless blood pressure measuring devices in the project of myocardial infarction and stroke screening and intervention of nations (MISSION) will be introduced with the emphasis on its integration with biomarkers and imaging markers for the early prediction of acute cardiovascular diseases.

Keywords: Optical and photonic sensors and systems, Physiological monitoring - Instrumentation, Physiological monitoring - Modeling and analysis
Abstract: MIT Laser Biomedical Research Center has pioneered and advanced noninvasive blood glucose monitoring using NIR Raman spectroscopy. Developed instruments have been used for in-vivo human studies and improved based on clinical feedback.

Keywords: Microfluidic applications
Abstract: Detecting the primary sequences for DNA and protein at single molecule level are at the frontier of genomics and proteomics research. In this presentation, I am going to talk about the single molecule DNA sequencing and then discuss the single molecule protein sequencing method.

Keywords: Micro- and nano-sensors, Electric fields - Tissue regeneration
Abstract: The solid state nanopore sensor as a high-throughput and low-cost technology can detect the enzyme catalytic reaction in a solution, it has ability to detect enzyme reaction in the nanoscale environment. In the present study, we used diameter ~28nm Si3N4 nanopore to detect the single horseradish peroxidase (HRP) molecule catalysis substrates and a real-time monitored enzyme catalysis substrates were investigated. This approach offers the potential for further development as studying gene expression, enzyme dynamics at the single-molecule level and quantitative test of small molecules.

Keywords: Cellular force transduction - Mechanical stimuli and mechanotransduction, Cellular force transduction - Chemistry and molecular mechanics, Cellular force transduction - Cell spreading and adhesion
Abstract: In this study, Molecular dynamics simulations were employed to investigate the roles of external ligand binding and internal ligand binding on β2 integrin outside-in activation. The results indicates that the internal ligand binding is critical for allosteric pathways from I domain to hybrid domain. These results further the understandings in the structure-function relationship of I-domain-containing integrins.

Keywords: Microfluidic techniques, methods and systems, Microfluidic applications, Biomaterials - Chemical and electrochemical sensors
Abstract: Protein machines work in aqueous solutions, and conformational flexibility is a key feature for proper functioning. Liquid phase electron microscopy enables molecular imaging with high spatial and temporal resolutions, which gives the opportunity to directly observe the conformational dynamics of single protein machines in their hydrated state. Here we provide preliminary studies on single protein imaging in solution at room temperature and discuss the strategies undertaken in our laboratory to improve imaging quality and resolution.

Keywords: Micro- and nano-sensors, Nano-bio technology design, Micro- and nano-technology
Abstract: Gag is the main viral structural protein that orchestrates the generation of retroviral particles. To drive particle assembly, thousands of Gag monomers must coalesce within a larger pool of host cellular proteins to form a highly-ordered Gag multimer at the plasma membrane (PM). To date, it is widely accepted that this process is mediated by Gag binding to long-stranded RNAs, including viral and cellular RNAs, which serve as scaffolds upon which Gag binds nonspecifically and then multimerizes. However, how Gag interacts with the scaffolding RNAs during the assembly process still remains an open question. We plan to employ super-resolution imaging techniques including photoactivated localization microscopy (PALM) and pair correlation PALM, combined with single-molecule FRET (smFRET), to map out the spatial organization, stoichiometry and dynamics of the interactions between Gag and RNAs at the single-molecule level. We hope these findings can provide new insights into the role of RNA in retroviral biogenesis in host cells, benefiting the design of novel therapeutics against retrovirus transmission and infectivity.

Keywords: Neurological disorders
Abstract: Freezing of Gait (FOG) is a highly debilitating and poorly understood symptom of Parkinson’s disease (PD), causing severe immobility and decreased quality of life. Turning Freezing (TF) is known as the most common sub-type of FOG, also causing the highest rate of falls in PD patients. During a TF, the feet of PD patients appear to become stuck whilst making a turn. This paper presents an electroencephalography (EEG) based classification method for detecting turning freezing episodes in six PD patients during Timed Up and Go Task experiments. Since EEG signals have a time-variant nature, time-frequency Stockwell Transform (S-Transform) techniques were used for feature extraction. The EEG sources were separated by means of independent component analysis using entropy bound minimization (ICA-EBM). The distinctive frequency-based features of selected independent components of EEG were extracted and classified using Bayesian Neural Networks. The classification demonstrated a high sensitivity of 84.2%, a specificity of 88.0% and an accuracy of 86.2% for detecting TF. These promising results pave the way for the development of a real-time device for detecting different sub-types of FOG during ambulation.

Keywords: Neurological disorders
Abstract: Gait initiation Failure (GIF) is the situation in which patients with Parkinson’s disease (PD) feel as if their feet get “stuck” to the floor when initiating their first steps. GIF is a sub-type of Freezing of Gait (FOG) and often leads to falls and related injuries. Understanding of neurobiological mechanisms underlying GIF has been limited by difficulties in eliciting and objectively characterizing such gait phenomena in the clinical setting. Studies investigating the effects of GIF on brain activity using EEG offer the potential to study such behavior. In this study, we present a novel methodology where wavelet transform was used for feature extraction and Support Vector Machine for classifying GIF events in five patients with PD and FOG. To deal with the large amount of EEG data, a Principal Component Analysis (PCA) was applied to reduce the data dimension from 15 EEG channels into 6 principal components (PCs), retaining 93% of the information. Independent Component Analysis using Entropy Bound Minimization (ICA-EBM) was applied to 6 PCs for source separation with the aim of improving detection ability of GIF events as compared to the normal initiation of gait (Good Starts). The results of this analysis demonstrated the correct identification of GIF episodes with an 83.1% sensitivity, 89.5% specificity and 86.3% accuracy. These results suggest that our proposed methodology is a promising non-invasive approach to improve GIF detection in PD and FOG.

Keywords: Neurological disorders - Stroke, Human performance
Abstract: Aim of this paper is to formulate a posturography stability score for stroke patients using fuzzy logic. Postural instability is one of the prominent symptoms of stroke, dementia, parkinsons disease, myopathy, etc. and is the major precursor of fall. Conventional scoring techniques used to assess postural stability require manual intervention and are dependent on live interaction with physiotherapist. We propose a novel scoring technique to calculate static stability of a person using posturography features acquired by Kinect sensor, which do not require any manual intervention or expert guidance, is cost effective and hence are ideal for tele rehabilitation purpose. Stability analysis is done during Single Limb Stance (SLS) exercise. Kinect sensor is used to calculate three features, naming SLS duration, vibration index, calculated from mean vibration of twenty joints and sway area of Centre of Mass (CoM). Based on the variation of these features, a fuzzy rule base is generated which calculates a static stability score. One way analysis of variance (Anova) between a group of stroke population and healthy individuals under study validates the reliability of the proposed scorer. Generated fuzzy score are comparable with standard stability scorer like Berg Balance scale and fall risk assessment tool like Johns Hopkins scale. Stability score, besides providing an index of overall stability can also be used as a fall predictability index.

Keywords: Neurological disorders - Diagnostic and evaluation techniques, Neural signal processing, Neurological disorders - Epilepsy
Abstract: Interictal epileptiform spikes are the key diagnostic biomarkers for epilepsy. The clinical gold standard of spike detection is visual inspection performed by neurologists. This is a tedious, time-consuming, and expert-centered process. The development of automated spike detection systems is necessary in order to provide a faster and more reliable diagnosis of epilepsy. In this paper, we propose an efficient template matching spike detector based on a combination of spike and background waveform templates. We generate a template library by clustering a collection of spikes and background waveforms extracted from a database of 50 patients with epilepsy. We benchmark the performance of five clustering techniques based on the receiver operating characteristic (ROC) curves. In addition, background templates are integrated with existing spike templates to improve the overall performance. The affinity propagation-based template matching system with a combination of spike and background templates is shown to outperform the other four conventional methods with the highest area-under-curve (AUC) of 0.953.

Keywords: Neurological disorders - Stroke, Brain physiology and modeling, Neurological disorders - Treatment methodologies
Abstract: Objective: The present study proposes a cerebral infarction detection algorithm based on heart rate variability (HRV). Methods: It has been reported that infarction affects HRV. Therefore, infarction could be detected at an acute stage by monitoring HRV. This study uses multivariate statistical process control (MSPC), which is a well-known anomaly monitoring method. HRV data shortly after infarction onsets are collected by using the middle cerebral artery occlusion (MCAO) model in rats. This study prepares 11 MCAO-operated rats and 11 sham-operated rats. Three sham-operated rats’ data are used for model construction of MSPC, and the other 19 rats’ data are used for its validation. Results: The sensitivity and specificity of the proposed algorithm were 82 % and 75 %, respectively. Conclusion: An infarction onset could be detected at an acute stage by monitoring HRV.

Keywords: Neurorehabilitation, Brain-computer/machine interface, Neurological disorders - Stroke
Abstract: Recent studies have demonstrated the efficacy of brain-machine interfaces (BMI) for motor rehabilitation after stroke, especially for those patients with severe paralysis. However, a cerebro-vascular accident can affect the brain in many different manners, and damage in various areas can lead to similar or equal motor deficits. The location of the insult influences the way the brain activates when moving or attempting to move a paralyzed limb. Since the essence of a rehabilitative BMI is to precisely decode motor commands from the brain, it is crucial to characterize how lesion location affects the measured signals and if and how it influences BMI performance. This paper compares the performances of an electroencephalography (EEG)-based movement intention decoder in two groups of severely paralyzed chronic stroke patients: 14 with subcortical lesions and 14 with mixed (i.e., cortical and subcortical) lesions. We show that the lesion location influences the performance of the BMI when decoding the movement attempts of the paretic arm. The obtained results underline the need of further developments for a better individualization of BMI-based rehabilitative therapies for stroke patients.

Keywords: Brain image analysis, Image segmentation, Image feature extraction
Abstract: Traumatic brain injury is a serious public health problem in the U.S. contributing to a large portion of permanent disability. However, its early management and treatment could limit the impact of the injury, save lives and reduce the burden of cost for patients as well as healthcare systems. Subdural hematoma is one of the most common types of TBI, which its visual detection and quantitative evaluation are time consuming and prone to error. In this study, we propose a fully automated machine learning based approach for 3D segmentation of convexity subdural hematomas. Textural, statistical and geometrical features of sample points from intracranial region are extracted based on head Computed Tomography (CT) images. Then, a tree bagger classifier is implemented to classify each pixel as hematoma or no-hematoma. Our method yields sensitivity, specificity and area under the receiver operating curve (AUC) of 85.02%, 73.74% and 0.87 respectively.

Keywords: Brain image analysis, Image classification, Image feature extraction
Abstract: For high-dimensional magnetic resonance imaging (MRI) data, many feature selection methods have been proposed to reduce feature dimension in the study of computer-aided Alzheimer’s disease (AD) diagnosis. This paper presents a compartmental sparse feature selection method used for AD identification. Based on the derived atlas-based regions-of-interest (ROIs) of brain, the proposed method partitioned the T1-weighted MRI data into several compartments. It performs feature selection and classification compartmentally according to the local feature dimension estimation and local feature selection using sparse principal component analysis (SPCA) method followed with elastic-net logistic regression (ENLR) classifier. Experimental results showed that the proposed method improves the classification performance for small ROIs with high computational efficiency.

Keywords: Brain image analysis, Magnetic resonance imaging - Contrast-enhanced dynamic MRI, Multiscale image analysis
Abstract: Studying tumor growth using mathematical mod-els from magnetic resonance (MR) images is an important ap-plication that is believed to play a major role in cancer treatment by predicting tumor evolution, quantifying the response to therapy, and treatment planning. Reaction diffusion is the most popular model because of its simplicity and consistency with the biological growth process. However, most of the current growth models focus on presurgical images and likely without treatment. In this paper, we propose a new reaction diffusion model to consider the chemotherapy and radiotherapy effects on the tumor growth modelling for patients with low grade glioma. The proposed model does not consider the tensor information from diffusion tensor imaging. Instead it uses a weighted parameter to promote higher diffusivity in white matter. The radiotherapy and chemotherapy effects are considered as a loss terms in the proposed model. The preliminary results of the proposed model on 2 MR images show that, our model can effectively simulate tumor growth with high accuracies when treatments are ad-ministrated to low grade glioma patients.

Keywords: Brain image analysis, Magnetic resonance imaging - MR neuroimaging, Magnetic resonance imaging - Contrast-enhanced dynamic MRI
Abstract: Gliomas are primary brain tumors arising from glial cells. Gliomas can be classified into different histopathologic grades according to World Health Oraganization (WHO) grading system which represents malignancy. In this paper, we present a method to predict the grades of Gliomas using Radiomics imaging features. MICCAI Brain Tumor Segmentation Challenge (BRATs 2015) training data, its segmentation ground truth and the ground truth labels were used for this work. 45 radiomics features based on histogram, shape and gray-level co-occurrence matrix (GLCM) were extracted from each FLAIR, T1, T1-Contrast, T2 image to quantify the property of Gliomas. Significant features among 180 features were selected through L1-norm regularization (LASSO). Based on LASSO coefficient and selected feature values, we computed a LASSO score and gliomas were classified into low-grade glimoa (LGG) or high-grade glimoa (HGG) through logistic regression. Classification result was validated by a 10-fold cross validation. Our method achieved accuracy of 0.8981, sensitivity of 0.8889, specificity of 0.9074, and area under the curve (AUC) = 0.8870.

Keywords: Brain image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: In our previous study, we suggested that the difference between tensor-based metrics in the anterior part of the right putamen between 21 and 18 months age groups associated with speech development during this ages. Here we used a correlational analysis between verbal scores and determinant of the Jacobian matrix to confirm our hypothesis. Significant correlations in the anterior part of the right putamen between verbal scores and surface metric were revealed in the 18 and 21 age groups.

Keywords: Neural stimulation, Neurological disorders - Diagnostic and evaluation techniques, Neurological disorders - Epilepsy
Abstract: This paper presents a novel brain mapping system for seizure onset region identification and control in treating epileptic seizure of patients. The proposed system consists of a closed-loop seizure control system-on-chip (SoC) and a control panel with graphical user interface. It can be used to identify the seizure onset brain regions and determine the most suitable parameters for a closed-loop cortical stimulation for epileptic patients.

Keywords: Neural interfaces - Implantable systems, Brain functional imaging, Brain functional imaging - Spatial-temporal dynamics
Abstract: In this paper, we report our recent results of implantable optoelectronic devices dedicated for measurement and control of neural functions. By the combination of custom designed CMOS chip and post processing techniques, optical sensor and light source functions are integrated with low invasiveness.

Keywords: Neural interfaces - Implantable systems, Sensory neuroprostheses, Neural stimulation
Abstract: Recently, we have been developing a multi-channel intra-cortical microstimulation system, which comprises a pair of intra- and extra-body devices equipped with wireless data communication and power transmission. The intra-body device can drive one or up to sixty four of a microampere-range 64-channel stimulator chip we developed. Thus, the system is expected to be applicable to various neural prostheses. However, in order to evaluate the operations of such a prosthetic device, animal studies in vivo would be inevitable. Thus, we utilized the in-vivo voltage-sensitive dye imaging technique to examine spatio-temporal neural responses to microstimulations generated by the wireless device system. The experimental results demonstrated a usefulness of our setup, in which quantitative relationships between the stimuli and the corresponding neural responses can be investigated.

Keywords: Neural signal processing, Neural stimulation, Neurological disorders
Abstract: The purpose of this presentation is to provide an overview of neural signal processing techniques for closed-loop neuromodulation. After illustrating overall structure of closed-loop neuromodulation systems, the techniques for the stimulus artifact removal will be explained, and the methods for neural state monitoring and biomarker extraction will be described. Finally, the current status of neuromodulation based on neural signal processing will be presented in detail. Closed-loop neuromodulation system automatically adjusts stimulation parameters based on the brain response in real time. Adequate tools for signal sensing and signal processing can be used to obtain meaningful biomarkers reflecting the state of neural systems. Especially, an appropriate neural signal processing technique can optimize the details of stimulation for effective treatment of target disease. Neural signal-based biomarkers reflecting the pathophysiological statuses of patients are essential for closed-loop neuromodulation, and they should be developed from an understanding of the relationship between clinical states and neural signals.

Keywords: Brain-computer/machine interface, Neural interfaces - Implantable systems, Motor neuroprostheses
Abstract: The brain-machine interface (BMI) is a new method for man-machine interfaces, which enables us to control machines and to communicate with others only using brain signals. In the present study, we fabricated a multi-channel neural recording ASIC chip for a fully implantable wireless BMI system. The chip has 32 neural recording channels and the input referred noise is below 3 µVpp. In addition, we developed a prototype fully implantable wireless BMI system containing the ASIC chips. To confirm the proper performance as neural recording device, the BMI system was applied to in vivo animal experiment.

Keywords: Sensory neuroprostheses - Visual, Neural stimulation, Neurorehabilitation
Abstract: Despite the rapid development of retinal prostheses, subjects implanted with retinal prosthesis have reported a low visual acuity of prosthetic vision. A possible approach for improving visual acuity is to utilize electric field modulation technology to produce virtual electrodes. This study designed a novel epiretinal stimulating electrode array and explored its feasibility of producing controllable virtual electrodes through computational modeling. According to the simulation results, our designed electrode array could produce virtual electrodes with the use of current steering stimulation strategy, and the site of virtual electrodes could be controlled two-dimensionally. This study may provide support for the application of virtual electrodes in epiretinal prosthesis in the future for increasing the visual acuity of prosthetic vision.

Keywords: General and theoretical informatics - Machine learning, General and theoretical informatics - Predictive analytics, General and theoretical informatics - Supervised learning method
Abstract: We present a fast, efficient method to predict future critical events for a patient. The prediction method is based on retrieving and leveraging similar waveform trajectories from a large medical database. Locality-sensitive hashing (LSH), our theoretical foundation, is a model-free, sub-linear time, approximate search method enabling a fast retrieval of a nearest neighbor set for a given query. We propose a new variant of LSH, namely Collision Frequency LSH (CFLSH), to further improve the prediction accuracy without sacrificing any speed. The key idea is that the more frequently an element and a query collide across multiple LSH hash tables, the more similar they are. Unlike the standard LSH which only utilizes the linear distance calculation, in CFLSH, the short-listing step from a pool of pre-selected candidates filtered by hash functions to the final nearest neighbor set relies upon the frequency of collision along with distance information. We evaluate CFLSH versus the standard LSH using the L1 and cosine distances, for predicting acute hypotensive episodes on arterial blood pressure time series data extracted from the MIMIC II database. Our results show that CFLSH for the L1 distance has a higher prediction accuracy and further accelerates the sub-linear querying time obtained by the standard LSH.

Keywords: General and theoretical informatics - Machine learning, General and theoretical informatics - Supervised learning method, Bioinformatics - Bioinformatics for health monitoring
Abstract: Sleep is a very important physiological phenomenon for recovery of physical and mental fatigue. Recently, there has been a lot of interest in the quality of sleep and the research is actively under way. In particular, it is important to have a repetitive and regular sleep cycle for good sleep. However, it takes a lot of time to determine sleep stages using physiological signals by experts. In this study, we constructed an optimized classifier based on normalized mutual information feature selection (NMIFS) and kernel based extreme learning machine (K-ELM), and total 4 sleep stages (Awake, weak sleep (stage1+stage2), deep sleep(stage3+stage4) and rapid eye movement (REM)) were automatically classified. As a results, the average of the accuracy obtained by proposed method (NMIFS+K-ELM) is 2~3% higher than that of simple method (K-ELM).

Keywords: General and theoretical informatics - Machine learning, General and theoretical informatics - Pattern recognition, Health Informatics - Knowledge discovery and management
Abstract: There is a huge demand to apply classification in medical analysis. A traditional classifier requires having training samples from each class. However, in reality, it is possible that the testing set may include classes that are not in the training set. This inevitably causes an issue: data from undefined class will be assigned to predefined classes. To tackle this, we propose a semi-supervised variational Gaussian mixture model to perform multi-class classification with novelty detection. Comparing to some popular novelty detection methods, we demonstrate that it gets better performance on a thyroid disease data, by generating the distribution of predefined classes and undefined class, without explicitly setting a threshold.

Keywords: General and theoretical informatics - Machine learning, General and theoretical informatics - Predictive analytics
Abstract: There is considerable interest in developing inexpensive, nonintrusive diagnostic tests for Alzheimer’s disease (AD), in the hope these tests will facilitate early diagnosis and support design of effective treatments. While tests based on blood-borne biomarkers such as microRNAs have shown promise, work to date indicates these tests often do not generalize well: diagnostic models derived for one patient group are not accurate when applied to a new cohort. This paper presents a novel ensemble-based transfer learning methodology which induces models that provide accurate diagnoses across distinct patient groups without retraining. The algorithm combines information from supervised ensemble learning and unsupervised ensemble clustering to enable robust transfer learning. The efficacy of the approach is illustrated through a case study involving microRNA-based AD diagnosis. In this test, we use our algorithm to learn a diagnostic model on data from one patient group and then apply the model to a different target group, obtaining diagnostic accuracy which actually exceeds that of a state-of-the-art model trained directly on the target group.

Keywords: General and theoretical informatics - Machine learning, General and theoretical informatics - Algorithms, General and theoretical informatics - Computational disease profiling
Abstract: There is growing awareness that user-generated social media content contains valuable health-related information and is more convenient to collect than typical health data. For example, Twitter has been employed to predict aggregate-level outcomes, such as regional rates of diabetes and child poverty, and to identify individual cases of depression and food poisoning. Models which make aggregate-level inferences can be induced from aggregate data, and consequently are straightforward to build. In contrast, learning models that produce individual-level (IL) predictions, which are more informative, usually requires a large number of difficult-to-acquire labeled IL examples. This paper presents a new machine learning method which achieves the best of both worlds, enabling IL models to be learned from aggregate labels. The algorithm makes predictions by combining unsupervised feature extraction, aggregate-based modeling, and optimal integration of aggregate-level and IL information. Two case studies illustrate how to learn health-relevant IL prediction models using only aggregate labels, and show that these models perform as well as state-of-the-art models trained on hundreds or thousands of labeled individuals.

Keywords: Health Informatics - Electronic health records, General and theoretical informatics - Predictive analytics, General and theoretical informatics - Machine learning
Abstract: Electronic medical claims (EMCs) can be used to accurately predict the occurrence of a variety of diseases, which can contribute to precise medical interventions. While there is a growing interest in the application of machine learning (ML) techniques to address clinical problems, the use of deep-learning in healthcare have just gained attention recently. Deep learning, such as deep neural network (DNN), has achieved impressive results in the areas of speech recognition, computer vision, and natural language processing in recent years. However, deep learning is often difficult to comprehend due to the complexities in its framework. Furthermore, this method has not yet been demonstrated to achieve a better performance comparing to other conventional ML algorithms in disease prediction tasks using EMCs. In this study, we utilize a large population-based EMC database of around 800,000 patients to compare DNN with three other ML approaches for predicting 5-year stroke occurrence. The result shows that DNN and gradient boosting decision tree (GBDT) can result in similarly high prediction accuracies that are better compared to logistic regression (LR) and support vector machine (SVM) approaches. Meanwhile, DNN achieves optimal results by using lesser amounts of patient data when comparing to GBDT method.

Keywords: Cardiovascular regulation - Baroreflex, Cardiovascular regulation - Autonomic nervous system, Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability
Abstract: The study proposes an extension of cardiac baroreflex (cBR) sequence analysis, traditionally performed over spontaneous fluctuations of heart period and systolic arterial pressure, to typify peripheral resistance baroreflex (prBR) from spontaneous variations of peripheral resistances and diastolic arterial pressure. The prBR baroreflex sensitivity (BRSprBR) and percentage of prBR sequences (SEQ%prBR) were computed along with analogous quantities assessed over cBR (i.e. BRScBR and SEQ%cBR). The cBR and prBR were typified in healthy subjects at rest (REST) and during light bicycle ergometer exercise at 10 percent of the maximal effort (EXE). Both cBR and prBR were affected by EXE: indeed, BRScBR and SEQ%prBR were significantly reduced. Moreover, while BRScBR and BRSprBR were not significantly associated, SEQ%cBR and SEQ%prBR were, and the correlation coefficient was positive. This study suggests that prBR can be typified from spontaneous variabilities along with the more traditional cBR, thus enlarging the possibility of monitoring human cardiovascular control mechanisms.

Keywords: Cardiovascular regulation - Heart rate variability, Cardiovascular regulation - Autonomic nervous system, Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability
Abstract: Delirium is an important syndrome in intensive care unit (ICU) patients, however, its characteristics are still unclear. Many evidences showed that this syndrome can be related to the autonomic instability. In this study, we aimed to investigate the possible alterations of autonomic nervous system (ANS) in delirium patients in ICU. Electrocardiography (ECG) of every ICU patient was measured during routine daily ICU care, and the data were gathered to evaluate the heart rate variability (HRV). HRV of total 60 patients were analyzed in time, frequency and non-linear domains. As a result, we found that heart rates of delirium patients were more variable and irregular than non-delirium patients. These findings may facilitate early detection and prevention of delirium in ICU.

Keywords: Cardiovascular regulation - Heart rate variability
Abstract: The innate variability of the heart rate contains both random and temporally scaled components. The objective of this study was to design a stochastic and mathematically integrative model (SIM) of the control of human heart rate that could replicate the cardiac rhythm during rest and two distinct states of exercise. Heartbeat-to-heartbeat interval (RRI) were recorded from 8 healthy subjects during rest and two exercise levels. Signal memory in RRI sequences was estimated with autocorrelation, and probability density functions (PDFs) were created by fitting polynomial curves to the normalized histograms of the RRI sequences. The SIM generated fictive RRI sequences by randomly selecting RRI values from a PDF and integrating the series using parameters derived from the autocorrelation analysis. Fractal-scaling in the model RRI sequences was quantified with detrended fluctuation analysis. After optimizing the model with the autocorrelation-based and PDF parameters, the SIM was able to produce fictive RRI sequences with significant fractal scaling (statistically assessed with surrogate analysis, p < 0.001), and with fractal-scaling characteristics similar to the original human data (p = 0.62). Increasing the length of the SIM-generated sequences did not alter the temporal scaling of the model sequences. In conclusion, this research demonstrated a stochastic and integrative model that can replicate the temporally correlated characteristics of the human heart rate.

Keywords: Cardiovascular and respiratory signal processing - Complexity in cardiovascular or respiratory signals, Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability, Cardiovascular regulation - Autonomic nervous system
Abstract: The assessment of cardiovascular control complexity as derived from spontaneous heart period (HP) fluctuations can be improved by exploiting a multivariate (MV) approach. This work proposes the assessment of a normalized complexity index (NCI) of HP variability according to a k-nearest-neighbor approach based on local predictability performed in a MV nonuniform embedding space. The method allows the selection of the past components of HP, systolic arterial pressure (SAP) and respiration (R) most useful for the prediction of HP fluctuations. The NCI derived from the MV approach (NCI_MV) was compared to a NCI computed via the same technique applied in a univariate (UV) embedding space (NCI_UV) formed exclusively by HP past samples. Indexes were computed in 130 patients undergoing coronary artery bypass graft (CABG) surgery before and after the induction of general anesthesia. Thirty-eight subjects developed atrial fibrillation (AF) after surgery, while the remaining ones did not (noAF, n=92). Both NCI_UV and NCI_MV could separate AF from noAF patients and revealed a larger complexity of the AF subjects. However, the statistical power of the NCI_MV was superior given that the probability of type I error was smaller than that of NCI_UV. The assessment of cardiac control complexity could improve risk stratification of patients at risk of developing AF after CABG surgery.

Keywords: Cardiovascular and respiratory signal processing - Cardiovascular signal processing, Cardiovascular regulation - Heart rate variability, Cardiovascular and respiratory system modeling - Cardiovascular-Respiratory Interactions
Abstract: Transcutaneous stimulation of the auricular branch of the vagus nerve (ABVN) has been proposed as a non-invasive alternative to vagus nerve stimulation (VNS). However, its cardiovagal effects are inconsistent across studies, likely due to inhomogeneity in the stimulation parameters. Here, we evaluate respiratory-gated ABVN stimulation (Respiratory-gated Auricular Vagal Afferent Nerve Stimulation, RAVANS), where the stimuli are delivered in 1 s bursts during the exhalation phase of respiration, thus mimicking the breathing-induced modulation of cardiac vagal activity. In this study, we present preliminary results from an ongoing single-arm, open label trial investigating the effects of different intensities of RAVANS in hypertensive subjects. We found that a mid-intensity RAVANS stimulation (rated as a 5 on a 0-10 scale) increases the cardiovagal tone and reduces the sympathetic tone during a paced breathing task. The present results could contribute to optimize RAVANS as a non-invasive, low-cost therapeutic intervention for hypertension.

Keywords: Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability, Cardiovascular and respiratory signal processing - Time-frequency, time-scale analysis of cardiorespiratory variability
Abstract: The entropy of heart rate variability is one of the main features characterizing the complexity of the cardiovascular system. In order to take into account the multiscale nature of cardiovascular regulation, it was proposed to evaluate entropy with a multiscale approach, based on the estimation of Sample Entropy on progressively coarse-grained series (Multiscale Sample Entropy, MSE). Aim of this work is to investigate two methodological aspects related to MSE of cardiovascular signals. The first aspect regards the tolerance below which a couple of points are considered similar in a given embedding dimension, in particular how the way the tolerance is set at each level of coarse graining influences the MSE estimates. The second aspect regards whether heart rate and blood pressure (BP) signals are characterized by different MSE structures. To investigate these aspects, we analyzed 65 continuous BP recordings of more than 90-minute duration in healthy volunteers sitting at rest, and applied MSE estimators to beat-by-beat series of systolic BP, diastolic BP and pulse interval (inverse of heart rate). Results indicate that the way the tolerance is set during coarse graining influences substantially the MSE profile of cardiovascular signals, modifying the relative level of their unpredictability.

Keywords: Image classification, Image feature extraction, Magnetic resonance imaging - Other organs
Abstract: Prostate cancer (CaP) is the second most diagnosed cancer in men all over the world. In the last decades, new imaging techniques based on Magnetic Resonance Imaging (MRI) have been developed improving diagnosis. In practise, diagnosis can be affected by multiple factors such as observer variability and visibility and complexity of the lesions. In this regard, computer-aided detection and computeraided diagnosis systems are being designed to help radiologists in their clinical practice. We propose a new computer-aided system taking advantage of all MRI modalities (i.e., T2-W-MRI, DCE-MRI, DWMRI, MRSI). This system has been extensively tested on a dataset which has been made publicly available.

Keywords: Image classification, Infra-red imaging
Abstract: Age Related Macular Degeneration (AMD) is the third leading cause of blindness and the first one in the elderly cite{kawasaki2010prevalence}. AMD usually causes central blindness. In this paper, we propose to detect AMD caused vision impairment from gaze data. Compared with the current methods, e.g., Amsler grid, Microperimetry and Preferential Hyperacuity Perimetry, to detect vision impairments, the proposed method has several advantages. 1) It does not require the patient to stare at a fixed position throughout the test. 2) It does not require the patient to orally or manually report / mark out the vision impairment. 3) It is easy to operate thus a trained nurse is capable of operating the test. We collect gaze data while the patient is performing fixation and smooth pursuit. Features describing the gaze properties are extracted and SVM with linear kernel is trained to detect AMD impaired vision. To implement the proposed method, we collected gaze data of 74 eyes of 57 patients, who are diagnosed as AMD patient by clinicians. Nidek Microperimetry is adopted as gold standard. 57 eyes with normal vision and 17 eyes with impaired vision (blind at more than half test points in Nidek test) are used for test. The result verifies the effectiveness of detecting vision impairment from gaze data.

Keywords: Image classification, Image feature extraction, Ultrasound imaging - Other organs
Abstract: This paper presents an automatic lumbar spine level identification system based on image processing of ultrasound images. The goal is to aid anesthetists in identifying the correct spinal level during epidural anesthesia. Spine level identification is initiated by detecting the location of the sacrum using a classifier based on a support vector machine. Image stitching is then conducted to produce a panorama image of the spinal area. During this process, the location of spinal processes are enhanced using a Gabor filter and detected through template matching. The locations of the spinal processes are tracked and used as an overlay on the ultrasound image in real-time. The system then informs the anesthetists when the correct spinal level has been reached. The system was evaluated on forty volunteers by two anesthetists with varying experience level and was able to detect the correct position of the L3-L4 spinal level in all of the volunteers. The average time taken to produce the location of the L3-L4 spinal level was 30.92 seconds. The results show that the system can quickly and accurately detect the location of the target spinal level.

Keywords: Image classification
Abstract: The demand for automatically recognizing medical images for screening, reference and management is growing faster than ever. Missing data phenomenon in medical image applications is common existence, and it could be inevitable. In this paper, we have addressed the problem of recognizing medical images with missing-features via Gaussian mixture model (GMM)-based approach. Since training a GMM by directly using high-dimensional feature vectors will result in instability, we have proposed a novel strategy to train the GMM from the corresponding reduced-dimensional one. The proposed method contains training and test phases. The former contains feature extraction, graph constrained nonnegative matrix factorization (NMF), GMM training, and the alternating expectation conditional maximization (AECM) for extending the reduced-dimensional GMM. In test phase, two methods, marginalizing GMM using Bayesian decision (MGBD) and conditional mean imputation (CMI), are applied to impute missing-features. Posterior probability of test images is calculated to identify objects. Experimental results on three real datasets demonstrate the feasibility and efficiency of the proposed scheme.

Keywords: Image classification, Image feature extraction
Abstract: Gastrointestinal (GI) bleeding detection plays an essential role in wireless capsule endoscopy (WCE) examination. In this paper, we present a new approach for WCE bleeding detection that combines handcrafted (HC) features and convolutional neural network (CNN) features. Compared with our previous work, a smaller-scale CNN architecture is constructed to lower the computational cost. In experiments, we show that the proposed strategy is highly capable when training data is limited, and yields comparable or better results than the latest methods.

Keywords: Image classification, Image feature extraction, Image segmentation
Abstract: We present a novel system for the automatic detection of angiodysplasia lesions from capsule endoscopy images. The approach identifies potential regions of interest and classifies them using a combination of color-based, texture, statistical and morphological features. A boosted decision tree classification method is used in order to overcome the problem of unbalanced sampling between pathological and non-pathological regions. The lesion detection method has been designed and validated using a lesion database labelled by an expert. The approach achieves a sensitivity of 89.51% and a specificity of 96.8%, thus providing a high performance in the detection of angiodysplasia lesions.

Keywords: General and theoretical informatics - Big data analytics, Bioinformatics - Cancer genomics, Neuro genomics, Cardio genomics, General and theoretical informatics - Social media
Abstract:

Keywords: Physiological systems modeling - Signal processing in physiological systems, Nonlinear dynamic analysis - Volterra-Wiener models in physiological systems, Time-frequency and time-scale analysis - Wavelets
Abstract: The performance of any intra-cardiac electrogram processing method is limited by the accuracy of its activation detection approach. The most common activation detection approaches in the literature aim to find the highest peak in the activation envelope disregarding the start and end points. However, the duration of the activation can be used to extract useful information such as wave collisions. In this work, we propose a novel orthogonal-based (FOS) approach for fast and accurate estimation of the start and end of the activations (activation envelope) in intracardiac recordings during atrial fibrillation. Wavelet decomposition of the signals was used to create a pool of basis functions for the proposed modeling method. The database included 24 recordings of approximate length of 6s obtained from inner walls of atria of 5 patients who underwent catheter ablation therapy. The start and end of activations in each electrogram was manually annotated by an expert electrophysiologist and the annotations were used as golden standard to calculate the performance of our envelope detection method. The results show promising performance and excellent robustness to training data for our proposed method with respect to envelope estimation error.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Physiological systems modeling - Signals and systems
Abstract: Cardiotocography (CTG) is the most common non-invasive diagnostic technique to evaluate fetal well-being. It consists in the recording of fetal heart rate (FHR; bpm) and maternal uterine contractions. Among the main parameters characterizing FHR, baseline (BL) is fundamental to determine fetal hypoxia and distress. In computerized applications, BL is typically computed as mean FHR±ΔFHR, with ΔFHR=8 bpm or ΔFHR=10 bpm, both values being experimentally fixed. In this context, the present work aims: to propose a statistical procedure for ΔFHR assessment; to quantitatively determine ΔFHR value by applying such procedure to clinical data; and to compare the statistically-determined ΔFHR value against the experimentally-determined ΔFHR values. To these aims, the 552 recordings of the “CTU-UHB intrapartum CTG database” from Physionet were submitted to an automatic procedure, which consisted in a FHR preprocessing phase and a statistical BL assessment. During preprocessing, FHR time series were divided into 20-min sliding windows, in which missing data were removed by linear interpolation. Only windows with a correction rate lower than 10% were further processed for BL assessment, according to which ΔFHR was computed as FHR standard deviation. Total number of accepted windows was 1192 (38.5%) over 383 recordings (69.4%) with at least an accepted window. Statistically-determined ΔFHR value was 9.7 bpm. Such value was statistically different from 8 bpm (P<10-19) but not from 10 bpm (P=0.16). Thus, ΔFHR=10 bpm is preferable over 8 bpm because both experimentally and statistically validated.

Keywords: Signal pattern classification, Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signal processing in physiological systems
Abstract: Dysphoric patients show symptoms associated with Major Depression, although within a narrowed symptomatology spectrum. In prevailing practice, clinicians assess Dysphoria through psychological scores and questionnaires exclusively, therefore without taking into account objective biomarkers. In this study, we investigated heartbeat linear and nonlinear dynamics aiming to an objective assessment of Dysphoria. To this end, we derived standard and nonlinear measures from heart rate variability (HRV) series gathered from dysphoric (n=14) and nondysphoric (n=17) undergraduate students during 5 minutes of resting state. We performed both statistical and pattern recognition analyses in order to discern the two groups. Results showed significant group-wise differences in HRV nonlinear metrics exclusively, suggesting a crucial role of nonlinear sympatho-vagal dynamics in Dysphoria. Furthermore, we achieved a classification accuracy of 77.52% for the automatic identification of Dysphoria at a single subject level.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Time-frequency and time-scale analysis - Wavelets, Signal pattern classification
Abstract: The aim of this study was to discriminate the electroencephalogram (EEG) background activity of 5 vascular dementia (VaD) patients, 15 stroke-related patients with mild cognitive impairment (MCI), and 15 control healthy subjects during a working memory (WM) task. We used independent component analysis (ICA) and wavelet transform (WT) as a hybrid pre-processing approach for EEG artifact removal. Three different features were extracted from the cleaned EEG signals: spectral entropy (SpecEn), permutation entropy (PerEn) and Tsallis entropy (TsEn). Two classification schemes were applied – support vector machine (SVM) and k-nearest neighbors (kNN) – with fuzzy neighborhood preserving analysis with QR-decomposition (FNPAQR) as a dimensionality reduction technique. The FNPAQR dimensionality reduction technique increased the SVM classification accuracy from 82.22% to 90.37% and from 82.6% to 86.67% for kNN. These results suggest that FNPAQR consistently improves the discrimination of VaD, stroke-related MCI patients and control healthy subjects and it could be a useful feature selection to help the identification of patients with VaD and stroke-related MCI.

Keywords: Time-frequency and time-scale analysis - Time-frequency analysis, Physiological systems modeling - Signal processing in physiological systems, Physiological systems modeling - Signal processing in simulation
Abstract: The aging process leads to diverse changes in the human organism, including in autonomic system modulation. In this study, we calculated indices of HRV in frequency (power spectral density, PSD) and time (the impulse response (IR) method) domains, using data from healthy young and elderly volunteers (Fantasia database from Physionet). The results obtained showed that aging leads to an attenuation of vagal modulation of elderly individuals when compared to young volunteers.

Keywords: Nonlinear dynamic analysis - Biomedical signals
Abstract: Alzheimer’s disease (AD) is a progressive and irreversible brain disorder affecting memory, thinking, and emotion. It is a prominent cause of dementia and a major societal problem worldwide. The complexity of brain recordings has been successfully used to help to characterize AD. We have recently introduced multiscale dispersion entropy (MDE) as a very fast and powerful tool to quantify the complexity of signals across several time scales. We now seek to assess the ability of MDE, in comparison with multiscale permutation entropy (MPE) and multiscale entropy (MSE), to discriminate resting-state magnetoencephalogram (MEG) recordings of 36 AD patients from 26 elderly age-matched control subjects. The results confirmed that MDE, unlike MSE, does not lead to undefined values, especially at high temporal scales. Moreover, the significant differences for AD patients versus controls obtained by MDE, in comparison with MSE and MPE, were at a larger number of scales. In addition, the computation time for our recently developed MDE was considerably less than that for the MSE and even the MPE.

Keywords: Signal pattern classification, Independent component analysis, Physiological systems modeling - Signal processing in physiological systems
Abstract: Patients suffering from neuro-degenerative diseases have difficulties with normal locomotion. This problem progresses with the course of disease. Gait assessment is an effective way of diagnosing the disease and quantifying its progress which can effectively prevent falls. In this paper, an automatic assessment method for analyzing gait data obtained by force sensor insoles is introduced. The gait analysis method is based on measuring the complexity of gait data after extracting independent sources. The results are promising an average accuracy of 94% for three different diseases.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Coupling and synchronization - Coherence in biomedical signal processing
Abstract: Analysis of brain signal complexity reveals the intrinsic network dynamics and is widely utilized in the investigation of mechanisms in mental disorders. In this study, the complexity of resting-state functional magnetic resonance imaging (fMRI) signals was explored in patients with depression using multiscale entropy (MSE). Thirty-five patients diagnosed with depression and 22 age- and gender-matched healthy controls were considered. The MSE profiles in 5 brain networks of the 2 participant groups were evaluated and analyzed. The results showed that depressive patients exhibited higher complexity in the left frontoparietal network than that seen in healthy controls, which is known to be critical for executive control functions. Through this study, the efficacy of MSE in identifying and understanding the mental disorders was also demonstrated.

Keywords: Nonlinear dynamic analysis - Nonlinear filtering, Nonlinear dynamic analysis - Biomedical signals
Abstract: Multiple-Choice Decision-Making Tasks are widely used to analyze behavior and infer underlying cognitive states that shape the decision and learning processes. The behavioral signals recorded in these tasks are dynamic and often non-Gaussian – for instance, when learning a multiple choice association task. Previously developed estimation algorithms for latent behavioral variables do not address multiple-choice responses. In this research, we use a state-space modeling framework to predict a cognitive learning state related to multiple choice decisions, which are best described by a multinomial distribution. The proposed algorithm combines a multinomial filter/smoother and a variational Bayes technique to estimate the dynamics of a learning state vector. The algorithm is applied to decision response data recorded from non-human primates (NHPs) performing a Multiple-Choice Decision Task.

Keywords: Nonlinear dynamic analysis - Nonlinear filtering, Principal component analysis
Abstract: Riemannian geometry has been found accurate and robust for classifying multidimensional data, for instance in brain-computer interfaces based on electroencephalography. Given a number of data points on the manifold of symmetric positive-definite matrices, it is often of interest to embed these points in a manifold of smaller dimension. This is necessary for large dimensions in order to preserve accuracy and useful in general to speed up computations. Geometry-aware methods try to accomplish this task while respecting as much as possible the geometry of the original data points. We provide a completely unsupervised closed-form solution for this problem and we show that it allows substantial dimensionality reduction without affecting the classification accuracy in two large public brain-computer interface data bases.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signal processing in physiological systems, Principal component analysis
Abstract: Physiological hand tremor causes undesirable vibration of hand-held surgical instruments which results in imprecisions and poor surgical outcomes. Existing tremor cancellation algorithms are based on detection of the tremulous motion from the whole motion; then adding an anti-phase tremor signal to the whole motion to it cancel out. These techniques are based on adaptive filtering algorithms which need a reference signal that is correlated with the actual tremor signal. Hence, adaptive algorithms use a non-linear phase filter to pre-filter the tremor signal either offline or in real-time. However, pre-filtering causes unnecessary delays and non-linear phase distortions as the filter has frequency selective delays. Consequently, the anti-phase tremor signal cannot be generated accurately which results in poor tremor cancellation. In this paper, we present a new technique based on singular spectrum analysis (SSA) and its recursive version, that is, recursive singular spectrum analysis (RSSA). These algorithms decompose the whole motion into dominant voluntary components corresponding to larger eigenvalues and oscillatory tremor components having smaller eigenvalues. By carefully selecting a group of specific decomposed signals based on their eigenvalues and spectral range, both voluntary and tremor signals can be reconstructed accurately. We tested the SSA and RSSA algorithms with real physiological tremor data that was recorded from five novice subjects. This new approach shows the tremor signal can be estimated from the whole motion with an accuracy of up to 85% offline. In real-time, tolerating a delay of ≈72ms, tremor signal can be estimated with at least 70% accuracy. This delay was found to be one-tenth of the delay caused by a conventional linear phase filter to achieve similar performance in real-time.

Keywords: Signal pattern classification, Data mining and processing in biosignals, Time-frequency and time-scale analysis - Time-frequency analysis
Abstract: Automated real time seizure detection is a challenging work since the sensitivity, false detection rate and seizure onset detection latency need to be considered simultaneously. Traditional pattern recognition and classification system with fixed channels and features usually suffers huge performance variation due to patient specificity and algorithm adaptability. To address this problem, we propose a two stage seizure detection system which has integrated channel selection and feature selection as an off-line preprocessing prior to final model is constructed. This system enables patient specific channel adjustment and flexible feature set extraction for each patient, thus a more compact and reliable model could be developed for patient customization. Employing the two stage scheme to the raw EEG data not only decreases the hardware cost in signal readout and feature extraction, but also remarkably improve the detection sensitivity and reduce false detections. Mutual information based method has been used for channel selection, while Random Forests and nonlinear SVM-RFE have been evaluated for feature selection. The whole system has achieved a mean detection latency of 6 seconds and false detection rate of 0.356 per hour. With two missed detections on the test database, the sensitivity is 74.2% by sample or 98.4% by record. Our design is also hardwarefriendly, which could be finally implemented in an on-chip closed loop neural modulation system.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Signal pattern classification
Abstract: Sport related mild traumatic brain injury (mTBI), generally known as a concussion, is a worldwide critical public health concern nowadays. Despite growing concern emphasized by scientific research and recent media presentation regarding mTBI and its effect in athletics life, the management, and prevention of mTBI are still not properly done. The evaluation mainly hampered due to the lack of proper knowledge, subjective nature of assessment tools including the fact that the brain functional deficits after mTBI can be mild or hidden. As a result, development of an effective tool for proper management of these mild incidents is a subject of active research. In this paper, to examine the neural substrates following mTBI, an analysis based on electroencephalogram (EEG) from twenty control and twenty concussed athletes is presented. Preliminary results suggest that the concussed athletes have a significant increase in delta, theta and alpha power but a decrease in beta power. We also calculated the power for individual frequencies from 1 Hz to 40 Hz in order to find out the specific frequencies with the highest deficits. The significant deficiencies were found at 1-2 Hz of delta band, 6-7 Hz of theta band, 8-10 Hz of the alpha band, and 16-18 Hz and 24-29 Hz of the beta band. Though there was no significant difference as observed in gamma band, we found the deficit was significant at 34-36 Hz range within the gamma band. The observed deficits at various frequencies demonstrate that even if there is no significant difference in the traditional frequency bands, there may be hidden deficits at some specific frequencies within a frequency band. These preliminary results suggest that the EEG analysis at each unity frequency may be more promising means of identifying the neuronal damage than the traditional frequency band based analysis. Eventually, the proposed analysis can provide an improved approximation to monitor the pathophysiological recovery after a concussion.

Keywords: Physiological systems modeling - Multivariate signal processing, Physiological systems modeling - Signal processing in physiological systems, Time-frequency and time-scale analysis - Nonstationary processing
Abstract: Epilepsy affects around 50 million people worldwide. Over 30% of patients are drug-resistant where the only treatment may be surgical resection of the epileptogenic zone (EZ), the region of the brain that generates seizures. Identification of the EZ is often based on invasive EEG recordings. As such, surgical outcome relies heavily on precise and dense placement of EEG electrodes into the brain. Despite large brain regions being removed, success rates barely reach 65%. This gives rise to the “missing electrode problem”, where clinicians want to know what neural activity looks like between sparsely implanted electrodes. Solving this problem will enable more accurate localization of the EZ. In this paper, we demonstrate the first steps towards developing a computational platform to estimate neural activity at the “missing electrodes” using a reduced-order observer from control theory. Specifically, we constructed a sequence of discrete time Linear Time-Invariant (LTI) models using the available EEG data from two epilepsy patients. Then, we used the models to simulate EEG data and remove selected signals (“missing” states) from the simulated data set. Finally, we used a reduced-order observer to estimate the signals of these “missing” states and evaluated performance by comparing the observer estimates to the simulated EEG time series.

Keywords: Physiological systems modeling - Signal processing in physiological systems
Abstract: Development of accurate fatigue level prediction models is of great importance for driving safety. In parallel, a limited number of sensors is a prerequisite for development of applicable wearable devices. Several EEG-based studies so far have performed classification in two or few levels, while others have proposed indices based on power ratios. Here, we utilized a regression Random Forest model in order to provide more accurate continuous fatigue level prediction. In detail, multiband power features were extracted from EEG data recorded from one hour simulated driving task. Next, cross-subject regression was performed to obtain common fatigue-related discriminative features. We achieved satisfactory prediction accuracy and simultaneously we minimized required electrodes, proposing to use a set of 3 electrodes.

Keywords: Signal pattern classification, Data mining and processing - Pattern recognition, Physiological systems modeling - Multivariate signal processing
Abstract: Brain decoding is essential in understanding of where and how information is encoded inside the brain. Existing literature has shown that a good classification accuracy is achievable in decoding for single subjects, but multi-subject classification has proven difficult due to the inter-subject variability.In this paper, multi-modal neuroimaging was used to improve two-class multi-subject classification accuracy in a cognitive task. In this transfer learning problem, a feature space based on special-form covariance matrices manipulated with riemannian geometry are used. A supervised two-layer hierarchical model was trained iteratively for estimating classification accuracies. Results are reported on a publically available multi-subject, multi-modal human neuroimaging dataset from MRC Cognition and Brain Sciences Unit, University of Cambridge. The dataset contains simultaneous recordings of electroencephalography (EEG) and magnetoencephalography (MEG). Our model attained, using leave-one-subject-out cross-validation, classification accuracy of 70.82% for single modal EEG, 81.55% for single modal MEG and 84.98% for multi-modal M/EEG.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Time-frequency and time-scale analysis - Time-frequency analysis, Physiological systems modeling - Signals and systems
Abstract: The taste is a vital sense in humans, because of its active role in regulating nutrition or avoiding harmful substances. Several studies showed the important role of the brain Pre-Frontal Cortex in decoding information coming from the gustatory system. It is also widely known, in neuroscientific literature, that the asymmetry of Pre-Frontal Cortex Activity is closely linked to the feeling of pleasantness experienced by the subject during sensorial stimulation. In this regard, from the electroencephalographic (EEG) signal it is possible to estimate the Approach/Withdrawal (AW) index, which has been largely investigated and validated in scientific literature, regarding visual, acoustic and olfactory stimuli. On the contrary there are no evidences about the relationship between such an index and taste. In this work, it has been investigated if such AW index could provide reliable information regarding the pleasantness felt during taste experience. In particular, the EEG-based AW indexes of 15 healthy subjects tasting 7 creams of different flavor have been compared with the subjective appreciation judgement. The results showed a high significant correlation between EEG-based and subjective data. Also, it has been used to statistically investigate differences of perception between the different flavors, demonstrating the potential applicability of such technology towards real applications.

Keywords: EEG imaging
Abstract: Resting state networks (RSNs) are human brain networks formed by spontaneous activity fluctuations in distributed brain regions when people are in task-free and awake state. RSNs have been so far extensively studied using functional magnetic resonance imaging (fMRI). Recently, electroencephalography (EEG) and magnetoencephalography (MEG) have also been used to derive RSNs, in which independent component analysis (ICA) is the key step. In these studies, ICA has been either directly applied to recorded data at sensors (sensor-space ICA) or estimated source data from sensors using inverse source imaging techniques (source-space ICA). Both sensor-space and source-space ICAs have demonstrated the capability in finding RSNs from EEG/MEG data and their results showed strong correlations to fMRI RSNs. However, their performance was hardly compared even differences have been observed in their results. In the present study, we compared the source-space and sensor-space ICAs in reconstructing spatial, temporal and spectral features of RSNs in both simulated and real EEG data. Results from simulated data indicated that the source-space ICA has better performance in reconstructing spatial, temporal, and spectral feature of RSNs. Results from resting-sate EEG data in seven healthy participants also showed the difference between two procedures and, through the comparison with RSN templates constructed from fMRI data, the source-space ICA indicated relatively better performance than the sensor-space ICA.

Keywords: Image classification, Ultrasound imaging - Other organs
Abstract: Computer-aided diagnosis (CAD) of liver cancers on contrast-enhanced ultrasound (CEUS) has attracted considerable attention in recent years. The enhancement patterns on CEUS for liver lesions consist of the arterial, portal venous and late phases. Several typical images selected from these three phases can provide reliable information basis for diagnosis of liver lesions. Therefore, we propose to develop a CAD framework for liver cancers with only one B-mode images and three typical CEUS images selected from three enhancement patterns, which simulates the clinical diagnosis mode of radiologists. Moreover, a framework of two-stage multi-view learning (TS-MVL) is proposed to perform both feature-level and classifier-level MVL for the diagnosis of liver cancers with multimodal ultrasound images. We propose to apply the nonlinear kernel matrix (NKM) algorithm to effectively fuse the features of multimodal ultrasound images, and then perform the multiple kernel boosting (MKB) algorithm to promote the predictive performance of multiple classifiers according to multi-view features. The experimental results indicate that the proposed algorithm outperforms the commonly used multi-view learning algorithms.

Keywords: Ultrasound imaging - Cardiac, Ultrasound imaging - Elastography, Cardiac imaging and image analysis
Abstract: Ultrasound myocardial elastography is a promising technique to estimate regional myocardial function. In this study, we proposed a fast mapping method to map myocardial elastogram. A nude mouse’s heart was scanned in supine position by an ultrasound system. The parasternal long-axis view of the heart and the ultrasound radio frequency (RF) signals were acquired for dynamic estimation of myocardial elasticity. The displacement and strain were calculated using analytic minimization (AM) and linear polynomial curve fitting method, respectively. The fast mapping method was proposed to map myocardial elastogram. The results display the contraction of myocardium intuitively. The method in this study is proved to have a potential to estimate viable myocardium in the future.

Keywords: Ultrasound imaging - Interventional, Image segmentation, Rigid-body image registration
Abstract: Laparoscopic pelvic lymph node dissection is a delicate operation because the pelvic arteries, which should be located firstly to guide the dissection, are often concealed by tissues and missed in the endoscopic view, so artery can be hurt when located inaccurately. In order to improve the safety and efficiency of the dissection, we developed an image-guided navigation system to provide the position information of the pelvic arteries for surgeons by registering the 3D artery model extracted from CT images to 3D model reconstructed from free-hand laparoscopic ultrasound images. The ultrasound probe is tracked using a proposed stereo vision based tracking strategy, which can simplify the system and reduce the setting time, and the artery is segmented from 2D ultrasound images using a local phase based snakes framework. The accuracy of navigation system using the stereo visual tracking was estimated using a phantom, and the TRE error was 1.58±0.70mm, and the feasibility of the proposed navigation system was confirmed in animal experiment.

Keywords: Ultrasound imaging - Cardiac, Image segmentation, Cardiac imaging and image analysis
Abstract: Echocardiography is an important tool to detect early evidence of mitral valve degradation associated with rheumatic heart disease. The segmentation and tracking of the Anterior Mitral Leaflet helps to quantify the morphologic valve anomalies, such as the leaflet thickening, shape and the mobility changes. The tracking of this leaflet throughout the cardiac cycle is still an open challenge in the research community. The widely used active contours segmentation framework fails when faced with large leaflet displacement. In this work, we propose the integration of optical flow in an open-ended active contour framework to address this difficulty. This additional information promotes solutions with contours next to high leaflet displacements, resulting in superior performance. The algorithm was tested on 9 fully annotated real clinical videos, acquired from the parasternal long axis view. The algorithm is compared with our previous work. Results show a clear improvement in situations where the leaflet exhibits large displacement or irregular shapes, with an average error of 4.5 pixels and a standard deviation of 2 pixels.

Keywords: Ultrasound imaging - Breast
Abstract: Regular examination of breasts may prevent and help to cure because breast cancer is treatable when it is detected early. Therefore, a breast cancer screening modality being sensitivity and cost-effective like ultrasonic imaging modality (US), is strongly required. In addition, the combination of a conventional US and its adjunct, Color Doppler has been proved for decreasing the rate of false-positive in breast cancer diagnosis. Thus, combination of these imaging modalities in a breast cancer segmentation would provide some benefits as well. An effective method for feature segmentation, active contour model has been widely utilized for decades. A crucial stage that affects the performance of active contour model is the initialization. This paper proposes a novel method for an automatic initialization of active contour model designed specifically for US-based imaging modalities. The method estimates an initial contour by utilizing the fusion of conventional US and Color Doppler. Examples and comparisons with three state-of-the-art automatic initialization methods are demonstrated to present the advantages of the proposed method. The evaluation results show high accuracy of initialization as well as fast convergence to features of interest.

Keywords: Optical imaging
Abstract: Abstract—Cone-beam X-ray luminescence optical tomography (CB-XLOT) plays an important role in in vivo small animal imaging study, which can non-invasively image three-dimensional (3-D) distribution of x-ray-excitable nanophosphors deeply embedded in imaged object. However, CB-XLOT suffers from a low spatial resolution due to the ill-posed nature of optical reconstruction. To alleviate the ill-posedness of reconstruction and improve the imaging performance of XLOT, in this paper, we propose an iterative weighted L1 minimization method which is achieved by incorporating YALL1 (Your algorithm for L1 norm problems). The physical phantom experiment was conducted to evaluate the performance of the proposed method, where a custom-made cone-beam XLOT system was used as the imaging platform. The experimental results indicate that by applying the proposed iterative weighted strategy to YALL1 method, the reconstruction performance of XLOT can be improved when compared with the conventional YALL1 method.

Keywords: Magnetic resonance imaging - Image reconstruction, Magnetic resonance imaging - Parallel MRI
Abstract: Conventional phase-contrast (PC) flow MRI provides dynamic information of blood flow by using electro-cardiogram (ECG)-synchronized cine acquisitions and respiration control. However, this often results in relatively low acquisition efficiency and is unable to assess blood flow variabilities. Recently, real-time imaging without ECG gating and respiration control has been a promising technique to overcome these limitations, however, it leads to a significantly more challenging reconstruction problem associated with highlyundersampled data. In this paper, we present a model-based imaging method, which integrates low-rank modeling with parallel imaging, to enable 4D real-time phase-contrast flow MRI without ECG gating and respiration control. The proposed method achieves real-time imaging at a spatial resolution of 2.4 mm, temporal resolution of 35.2 ms, with three directional flow encodings, and well resolves beat-by-beat flow variations, which cannot be achieved by the conventional cine-based method. The proposed has been evaluated by in vivo data with multiple healthy subjects and one arrhythmic patient. For the first time, we demonstrate the feasibility of 4D real-time flow MRI.

Keywords: Magnetic resonance imaging - Image reconstruction, Image reconstruction - Fast algorithms, Image reconstruction and enhancement - Parametric image reconstruction
Abstract: A super-fast dictionary generation and searching (DGS) algorithm was developed for MR parameter quantification using magnetic resonance fingerprinting (MRF). MRF is a new technique for simultaneously quantifying multiple MR parameters using one temporally resolved MR scan. But it has a multiplicative computation complexity, resulting in a big burden of dictionary generating, saving, and retrieving, which can easily be intractable for any state-of-art computers. Based on retrospective analysis of the dictionary matching object function, a multi-scale ZOOM like DGS algorithm, dubbed as MRF-ZOOM, was proposed. MRF ZOOM is quasi- parameter-separable so the multiplicative computation complexity is broken into additive one. Evaluations showed that MRF ZOOM was hundreds or thousands of times faster than the original MRF parameter quantification method even without counting the dictionary generation time in. Using real data, it yielded nearly the same results as produced by the original method. MRF ZOOM provides a super-fast solution for MR parameter quantification.

Keywords: Image reconstruction and enhancement - Compressive sensing/sampling, Magnetic resonance imaging - Parallel MRI, Magnetic resonance imaging - Interventional MRI
Abstract: Recently, susceptibility based positive contrast MRI technique emerged as an effective method of visualizing the small MR compatible devices, such as brachytherapy seeds. One of the challenges associated with this method is the long scan time. In this work, we present an accelerated susceptibility based positive contrast MR imaging method, in which the susceptibility map can be generated from an under-sampled data. We use a combination of parallel imaging (GRAPPA) and compressive sensing (CS) technique in a hybrid k-space. The results in brachytherapy seeds imaging show that 3-D high-quality images can be achieved at an acceleration factor up to 4, which can decrease the scan time from 4 minutes to 1.4 minutes.

Keywords: Magnetic resonance imaging - Image reconstruction, Image reconstruction and enhancement - Compressive sensing/sampling, Magnetic resonance imaging - Parallel MRI
Abstract: Magnetic resonance fingerprinting (MRF) is a new quantitative imaging paradigm that enables simultaneous acquisition of multiple magnetic resonance tissue parameters (e.g., T1, T2, and spin density). Recently, MRF has been integrated with simultaneous multislice (SMS) acquisitions to enable volumetric imaging with faster scan time. In this paper, we present a new image reconstruction method based on low-rank and subspace modeling for improved SMS-MRF. Here the low-rank model exploits strong spatiotemporal correlation among contrast-weighted images, while the subspace model captures the temporal evolution of magnetization dynamics. With the proposed model, the image reconstruction problem is formulated as a convex optimization problem, for which we develop an algorithm based on variable splitting and the alternating direction method of multipliers. The performance of the proposed method has been evaluated by numerical experiments, and the results demonstrate that the proposed method leads to improved accuracy over the conventional approach. Practically, the proposed method has a potential to allow for a 3x speedup with minimal reconstruction error, resulting in less than 5 sec imaging time per slice.

Keywords: Magnetic resonance imaging - MR spectroscopy, Image reconstruction and enhancement - Compressive sensing/sampling, Image reconstruction and enhancement - Parametric image reconstruction
Abstract: Magnetic resonance spectroscopy has many important applications in bio-engineering while acquiring high dimensional spectroscopy is usually time consuming. Non-uniformly sampling can speed up the data acquisition but the missing data points have to be restored with proper signal models. In this work, a specific two dimensional (2D) magnetic resonance signal, in which the first dimension lies in frequency domain while the second dimension lies in time domain, is reconstructed with a proposed low rank Hankel matrix method. This method explores two general properties: 1) the rank of a structured matrix, converted from a 2D exponential signal, is equal to the number of 2D spectral peaks; 2) this rank is small if the spectrum is sparse. Result on realistic magnetic resonance spectroscopy shows that proposed method outperforms the state-of-the-art compressed sensing method on recovering low intensities spectral peaks.

Keywords: Implantable sensors, Physiological monitoring - Modeling and analysis, Wearable antennas and in-body communications
Abstract: Wireless capsule endoscopy (WCE) is a non-invasive technology used for inspection of the gastrointestinal tract. Localization of the capsule is a vital part of the system enabling physicians to identify the position of anomalies. Due to intestinal motility, the capsule positions will change with time, but the distance travelled in the intestine remains more or less constant. In this paper a method for calculating the pathlength travelled by a WCE is proposed. The method is based on Kalman- and particle filters. The travelled distance was estimated to an accuracy within a few millimeters.

Keywords: Wearable wireless sensors, motes and systems
Abstract: Recently, wearable vital sensor has been extensively studied and been installed into market. Most of the examples use Bluetooth as a data transmission scheme. European Telecommunication Standard Institute (ETSI) has been standardizing a new BAN standard called SmartBAN aiming at an ultra-low power consumption and co-existence friendly system. In this paper, we present the latest update of the SmartBAN standardization.

Keywords: Implantable technologies, Bio-electric sensors - Sensor systems
Abstract: To achieve a better understanding of which cortical excitability is linked to target brain area, we described the effects of noninvasive brain stimulation with transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) and investigated the way to improve the stimulus efficiency in terms of focality and intensity by introducing an implantable transcranial channel. We presented a multi-scale computational model that combines a volume conductor model of the head and multi-compartmental models of cortical neurons to explain the activation of pyramidal neurons in motor cortex. We then simulated neural activations as a function of coil orientation produced by TMS and neural polarizations during tDCS.

Keywords: Wearable power and on-body energy harvesting, Wearable body-compliant, flexible and printed electronics
Abstract: Wearable sensor nodes have gained a lot of attention during the past few years as they can monitor and record people's physical parameters in real time. Wearable sensor nodes can promote healthy lifestyles and prevent the occurrence of potential illness or injuries. This paper presents a flexible wearable sensor system powered by an efficient solar energy harvesting technique. It can measure the subject's heartbeats using a photoplethysmography (PPG) sensor and perform activity monitoring using an accelerometer. The solar energy harvester adopts an output current based maximum power point tracking (MPPT) algorithm, which controls the solar panel to operate within its high output power range. The power consumption of the flexible sensor nodes has been investigated under different operation conditions. Experimental results demonstrate that wearable sensor nodes can work for more than 12 hours when they are powered by the solar energy harvester for 3 hours in the bright sunlight.

Keywords: Image classification, Image feature extraction
Abstract: Colonoscopy is a standard method for screening polyps by highly trained physicians. Miss-detected polyps in colonoscopy are potential risk factor for colorectal cancer. In this study, we investigate an automatic polyp classification framework. We aim to compare two different approaches named hand-craft feature method and convolutional neural network (CNN) based deep learning method. Combined shape and color features are used for hand craft feature extraction and support vector machine (SVM) method is adopted for classification. For CNN approach, three convolution and pooling based deep learning framework is used for classification purpose. The proposed framework is evaluated using three public polyp databases. From the experimental results, we have shown that the CNN based deep learning framework shows better classification performance than the hand-craft feature based methods. It achieves over 90% of classification accuracy, sensitivity, specificity and precision.

Keywords: Physical sensors and sensor systems - New sensing techniques, Physiological monitoring - Instrumentation, Physiological monitoring - Novel methods
Abstract: Autonomous wearable biomedical sensors enable continuous human body vital signs monitoring with features such as being conformable, mobile, cost-effective and self- powered. In this work, we report on an autonomous and multi-positional sensor capable of heart rate and blood pressure monitoring. The device concept is based on a piezoelectric-charge-gated thin-film transistor (PCGTFT) where a polyvinylidene fluoride (PVDF) piezoelectric sandwich structure is incorporated with an amorphous silicon (a- Si:H) dual-gate TFT (DGTFT). An analytical model and preliminary experimental results will be presented along with a demonstration of a proof-of-concept sensing system for continuous multi-point heart rate monitoring.

Keywords: Physiological monitoring - Instrumentation, Wearable low power, wireless sensing methods
Abstract: Pulse transit time (ptt) is a widely researched approach for wearable, unobtrusive, blood pressure monitoring. The estimation of ptt, being a delay measurement, may be affected by the group delay introduced by the signal conditioning chain. In this work a previously reported method for estimating central (aortic) blood pressure from ptt at aortic domain, using ECG R wave and BCG J wave detection, is considered. A simple design approach for the signal conditioning chain, which is suitable for a wearable device and that takes care of minimizing the impact of eventually introduced spurious delays is presented. The design provides less than 2ms group-delay. The design of a wearable device prototype for ECG, BCG and ptt acquisition and experimental results of its application are reported, showing the effectiveness of the proposed approach.

Keywords: Physical sensors and sensor systems - New sensing techniques, Physical sensors and sensor systems - Optical and photonic Sensors and systems, Wearable sensor systems - User centered design and applications
Abstract: This research proposed a novel method to estimate the 3D contact force of a fingertip using a photoplethysmogram (PPG) device on the proximal part of a finger. The proposed method detects non-pulsatile and pulsatile components of PPG signals, extracts eight feature values related to the contact force, and estimates the force by multiple linear regression. In the validation experiments, the participants wore a PPG device at the proximal part of the index finger and applied contact force. Then, the relationship between the contact force and PPG feature values was investigated. As a result, the system was found to be capable of estimating the 3D contact force with a root-mean-square error of 3.5 % of the maximum contact force.

Keywords: Physical sensors and sensor systems - Mechanical sensors and systems, Integrated wearable and portable systems, Wearable sensor systems - User centered design and applications
Abstract: We present the development and basic evaluation of a new wearable device for estimation of ankle joint impedance called Tarsusmeter. The device is intended for application with persons with locomotion disabilities to quantify the ankle joint impedance, especially in cases of spasticity where the joint's impedance is expected to differ significantly from healthy persons. The lack of a simple and light weight solution to provide objective evaluation of ankle joint impedance motivates the design criteria of this device to be as such. The target application is also to quantify variable stiffness actuator based orthosis in-vivo. Thus the form factor avoids overlap with custom shapes of such orthosis. The paper presents the mechanical design of the device, physical simulations to characterize the device-leg system, the used algorithm for impedance parameter estimation, and preliminary testing of the device.

Keywords: Physical sensors and sensor systems - Optical and photonic Sensors and systems, Physiological monitoring - Instrumentation, Portable miniaturized systems
Abstract: Millions worldwide suffer from diseases that exhibit early warnings signs that can be detected by standard clinical-grade diagnostic tools. Unfortunately, such tools are often prohibitively expensive to the developing world leading to inadequate healthcare and high mortality rates. To address this problem, a smartphone-based pulse oximeter is presented that interfaces with the phone through the audio jack, enabling point-of-care measurements of heart rate (HR) and oxygen saturation (SpO2). The device is designed to utilize existing phone resources (e.g., the processor, battery, and memory) resulting in a more portable and inexpensive diagnostic tool than standalone equivalents. By adaptively tuning the LED driving signal, the device is less dependent on phone-specific audio jack properties than prior audio jack-based work making it universally compatible with all smartphones. We demonstrate that the pulse oximeter can adaptively optimize the signal-to-noise ratio (SNR) within the power constraints of a mobile phone (< 10mW) while maintaining high accuracy (HR error < 3.4% and SpO2 error < 3.7%) against a clinical grade instrument.

Keywords: Physiological monitoring - Instrumentation, Physical sensors and sensor systems - Mechanical sensors and systems, Mechanical sensors and systems
Abstract: In this study, we proposed a measurement system that extracts heartbeat interval data from multiple piezoelectric sensors placed on a chair that eliminates noise generated by body movement. We asked five healthy males (21−24 years old) to sit in an arbitrary position on a chair that had eight piezoelectric sensors attached, and heartbeat signals were measured. The experiment consisted of four measurements (5 min measurements after adequate rest), while performing specific body movements at a specific time, with deliberately mixed noise of body movements. To remove body movement noise, bandpass filter processing (4 Hz−20 Hz) was applied to the signal obtained from the piezoelectric sensors, and the heartbeat component was extracted using independent component analysis (number of separations was eight) on the processed waveform. For verification, the error rate was obtained before and after the removal of body movement noise, respectively. The error rate after removal of body motion was 2.91 ± 0.75%. In our previous study with two piezoelectric sensors without body movement, the error rate was 2.47 ± 2.66%. Therefore, our proposed measurement system may improve the accuracy of heartbeat interval detection, which can be observed continuously by removing the body motion.

Keywords: Micro- and nano-sensors, Nano-bio technology design, Micro- and nano-technology
Abstract: A novel approach based on the nanoplasmon ruler has been established to achieve direct visualization of the dynamic intercellular communication process in the immune system. Such a novel approach will establish a new paradigm that permits, for the first time, the direct visualization of the dynamic intercellular communication process in the immune system. The knowledge obtained from this study will facilitate a more comprehensive understanding of the immune intercellular network, unlocking the potential to transform the experimental studies into an information-rich science not only in immunology but beyond.

Keywords: Microfluidic applications, Biomaterial-cell interactions - Engineered vascular tissue, Cellular force transduction - Mechanical stimuli and mechanotransduction
Abstract: As the essential parts of circulatory systems, vascular networks have appealed numerous attention in current clinical and translational researches. Engineering vascularized microenvironment models with tunable, multifunctional, and controllable capabilities are important for mimicking in vivo tissue conditions with the controllability in biochemical (e.g. growth factors, cytokines), biophysical (e.g. flow stress, substrate stiffness), intercellular communications (e.g. immune cells, cancer cells) and cell-matrix interactions. We developed a three-dimensional (3D) microfluidic vascularized microsystem integrated with tunable biochemical and biomechanical conditions as well as cell-cell interactions. Using such platform, we investigated the inflammation-biased angiogenesis in a controllable matrix mechanical property, adhesive signature, and biochemical conditions.

Keywords: Microfluidic applications, Microfluidic techniques, methods and systems
Abstract: Peritoneal dialysis is a treatment for patients who suffer from severe chronic kidney disease. To prevent any infection during the treatment, it is important to monitor the population of white blood cell in dialysis fluid. However, current cell counting techniques usually suffer from labor-intensive manipulation steps and cannot deal with low cell counts. Here, we develop a microfluidic device with hydrodynamic trap arrays to capture neutrophils conjugated with 30μm polystyrene beads and total white blood cell numbers. This microfluidic platform enables simultaneously cell trapping and selection without complicated sample processing steps and equipment. We believe this platform can provide real-time infection status of patient for timely treatment.

Keywords: Microfluidic applications, Microfluidic techniques, methods and systems
Abstract: Microfluidics, as popularized by the pioneering works of George Whitesides's group at Harvard, has led to many innovative solutions to biomedical applications. The recent advances by groups, such as Prof. Donald Ingber's at Harvard, have shown the possibility to mimic organs in microfluidic chips. In this talk, we will introduce features of a microfluidic chip that we constructed for mimicking the eye chamber. We demonstrate how the chip can be used for modeling the emulsification of oil in the eye following the surgical procedures for treating a type of retinal detachment. I will also discuss some of the recent unpublished works from our group in mimicking the retina.

Keywords: Microfluidic applications, Microfluidic techniques, methods and systems, Micro- and nano-technology
Abstract: This talk highlights three different microfluidic platforms for innovative and effective single cell studies and analysis: (1) a microfluidic system for spatiotemporally controlling the delivery of bio-reagents to cells cultured in a microfluidic channel, (2) a microfluidic quadrupole (MQ) formed within an open microfluidic chip to generate floating concentration gradients for cell chemotaxis studies, and (3) a microfluidic platform for isolating circulating plasma cells from multiple myeloma patients’ blood samples.

Keywords: Microfluidic applications, Micro- and nano-sensors, Microfluidic techniques, methods and systems
Abstract: We report a microfluidic immunoassay consisting of multiple precise and efficient mixing units for transient multi-cytokine detection secreted by lymphocytes. This device has an array of detection micro-chambers with each included a Taylor dispersion-based mixing unit. The sub-pico-liter bio-samples are extracted from cell culture media and mixed with cytokine-sensitive fluorescence micro-beads, which offer fluorescent signals to quantify the bound cytokines on beads, are embedded in the chambers for quantitative detection of the target cytokines. We optimize the mixing scheme on its robustness and sensitivity. We demonstrate the high detection sensitivity for the detection of multiple cytokine concentrations released by a human leukemic cell line (THP-1). Further development of this device can lead to the profiling of multiple cytokine dynamics of lymphocytes for deeper understandings of the human immune system as well as the immune diseases.

Keywords: Nano-bio technology design, Microfluidic applications, Micro- and nano-technology
Abstract: Traditional drug-discovery workflows are resource intensive and characterized with long timeliness. Advances in bioprocess technologies, based on automation and miniaturization, have been established to incrementally address these matters – however these technologies are dedicated to specific processes limiting their holistic application. In collaboration with Berkeley Lights Inc. (BLI), Amgen has worked on an innovative nanofluidic tool that couples cellular manipulation via Opto-Electro-Positioning (OEP), fluorescent detection, and software-driven automated workflow execution. The continued development and implementation of this platform will overcome the challenges currently associated with microfluidic, fragmented workflows in biopharma and life science research. The miniaturized automated nanofluidic workflows are transformational and allow massively parallel manipulation of individual cells and reagents.

Keywords: Neural stimulation, Neurological disorders, Neurorehabilitation
Abstract: In this presentation of the minisymposium, a non-invasive technique for suppressing tremor in patients with Parkinson’s disease (PD) by evoked cutaneous afferents using surface electrical stimulation will be discussed. The presentation will introduce the recent findings of spinal neural pathway for tremor signals, and on how the peripheral neuromuscular system interacts with descending tremor signals to generate tremor at limbs. These results led to the proposal and test of a hypothesis that cutaneous afferents can interfere the spinal passage of tremor signals at the propriospinal neuron network, thus causing suppression of tremor amplitude. Here, we report the results obtained in 8 PD patients, which support the above hypothesis.

Keywords: Neural stimulation - Deep brain, Neurological disorders - Mechanisms, Human performance - Speech
Abstract: Our objective is to study whether and how 60Hz stimulation, compared with 130Hz, affects vocal responses of patients with Parkinson’s Disease (PD). Patients vocalized for 3 seconds with their sound played back to ears in real-time. Individuals with PD, who also received bilateral DBS at STN, sustained a vowel sound (/a/) and received unexpected perturbations in voice loudness auditory feedback. Our results suggest that low-frequency (60Hz) stimulation is likely to differentially affect the vocal response depending on whether the perturbation was augmenting (upshift) or diminishing (downshift) the magnitude of loudness.

Keywords: Brain functional imaging - Segmentation
Abstract: Deep brain stimulation (DBS) is the only technique that can directly modulate brain activities using electrical stimulation. A vast amount of research has shown that DBS not only causes a direct effect on the target nuclei, but somehow modulates pathological oscillations that reverberate through multiple brain regions. As such, this provides great opportunities for understanding brain activity in human behaviors. In this talk, the progress of the research and development of DBS in China will be presented, including the device development, applications and innovations. In the last decade, High frequency stimulation (HFS) has been used for the treatment of advanced Parkinson’s disease (PD), which was honored by the Lasker Medical Research Award 2014. However, it is known that HFS is unable to alleviate certain symptoms, such as freezing of gait and dysarthria. We have recently found that the combination of different frequencies can successfully treat these unmanageable symptoms, using our newly developed stimulation therapy called ‘Various Frequency Stimulation (VFS)’. Several cases will be presented in this talk to show the significant differences of VFS to the currently used HFS. From HFS to VFS, a small change but a great deal of difference. There is an enormous challenge in understanding how to modulate the brain? One fact remains is that we need tools and technology. Based on many years’ experience on developing implantable devices, we have established a new platform that not only provides DBS therapy but also gives concurrent measurements of local field potential signals to investigate neural circuits. This research tool is like an implanted brain recorder that may allow us to answer vital questions related with various brain disorders and different brain states in combination with behavioral evaluations. In addition, this device also has great potential for providing high clinical value, such as guiding DBS parameter programming, and finally realizing the in

Keywords: Functional image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: It has recently become evident that the functional connectome of the human brain is a dynamical entity whose time evolution carries important information underpinning physiological brain function as well as its disease-related aberrations. While simple sliding window approaches have had some success in estimating dynamical brain connectivity in a functional MRI (fMRI) context, these methods suffer from limitations related to the arbitrary choice of window length and limited time resolution.Recently, Generalized autoregressive conditional heteroscedastic (GARCH) models have been employed to generate dynamical covariance models which can be applied to fMRI. Here, we employ a GARCH-based method (dynamic conditional correlation - DCC) to estimate dynamical brain connectivity in the Human Connectome Project (HCP) dataset and study how the dynamic functional connectivity behaviors related to personality as described by the five-factor model. Openness, a trait related to curiosity and creativity, is the only trait associated with significant differences in the amount of time-variability (but not in absolute median connectivity) of several inter-networkfunctional connections in the human brain. The DCC method offers a novel window to extract dynamical information which can aid in elucidating the neurophysiological underpinning of phenomena to which conventional static brain connectivity estimates are insensitive.

Keywords: Image reconstruction - Performance evaluation, Image visualization, Brain image analysis
Abstract: Mapping the microvascular networks in the brain can lead to significant scientific and clinical insights. We developed a serial sectioning microscopy technique called the Knife-Edge Scanning Microscopy (KESM) to section and image the entire mouse brain at submicrometer resolution.In our effort to map the entire vascular network in the mouse brain, we perfused the vessels with India ink and used KESM to image the prepared brain. This results in about 1.5 TB of raw image data which poses a serious challenge in terms of analysis. In this paper, we will present our data set, and computational algorithms we developed to trace and analyze morphological properties of the mouse brain vascular network. Since the data is available across the entire brain in full detail (the smallest capillaries can be observed in our data), it enables the comparison of regional differences in morphological properties. We expect our results to provide rich insights to brain and neuroscience researchers.

Keywords: Brain image analysis, Functional image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: Recently, the field of functional brain connectivity has shifted its attention on studying how functional connectivity (FC) between remote regions changes over time. It is becoming increasingly evident that the human "connectome" is a dynamical entity whose variations are effected over very short timescales and reflect crucial mechanisms which underline the physiological functioning of the brain. In this study, we employ ad-hoc statistical and surrogate data generation methods to quantify whether and which brain networks displayed dynamic behaviors in a very large sample of healthy subjects provided by the Human Connectome Project (HCP). Our findings provided evidences that there are specific pairs of networks and specific networks within the healthy brain that are more likely to display dynamic behaviors. This new set of findings supports the notion that studying the time-variant connectivity in the brain could reveal useful and important properties about brain functioning in health and disease.

Keywords: Functional image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: While estimates of complex heartbeat dynamics have provided effective prognostic and diagnostic markers for a wide range of pathologies, brain correlates of complex cardiac measures in general and of complex sympatho-vagal dynamics in particular are still unknown. In this study we combine resting state functional Magnetic Resonance Imaging (fMRI) and physiological signal acquisition from 34 healthy subjects selected from the Human Connectome Project (HCP) repository with inhomogeneous point-process approximate and sample heartbeat entropy measures (ipApEn and ipSampEn) to investigate brain areas involved in complex cardiovascular control. Our results show that activity in the Temporal Gyrus, Frontal Orbital Cortex, Temporal Fusiform and Opercular cortices, Planum Temporale, and Paracingulate cortex are negatively correlated with ipApEn dynamics. Activity in the same cortical areas as well as in the Temporal Fusiform cortex are negatively correlated with ipSampEn dynamics. No significant positive correlations were found. These pioneering results suggest that cardiovascular complexity at rest is linked to a few specific cortical brain structures, including crucial areas connected with parasympathetic outflow. This corroborates the hypothesis of a multidimensional central network which controls nonlinear cardiac dynamics under a predominantly vagally-driven tone.

Keywords: Magnetic resonance imaging - Diffusion tensor and diffusion spectrum imaging, Image feature extraction, Brain image analysis
Abstract: Effective measurement of cognitive impairment caused by Alzheimer’s disease (AD) will provide a chance for early medical intervention and delay the disease onset. Diffusion tensor imaging (DTI) provides a non-intrusive examination of cranial nerve diseases which can help us observe the microstructure of neuron fibers. Cognitive control network (CCN) consists of the brain regions that highly related to human self-control. In this study, hub-and-spoke model, which was widely used in transportation and sociology area, was proposed and introduced to analyze the interrelationship of CCN and other regions under its control. At meanwhile, cognitive control related network (CCRN) was built by applying this model. Local and global graph theoretical parameters were calculated and analyzed through statistical method. Significant difference had been found in the scale of local as well as global which may represent the impairment of cognitive control ability. This result may provide a potential bio-marker for the loss of connection caused by Alzheimer’s disease.

Keywords: Functional image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: A prominent pathway of brain-heart interaction is represented by autonomic nervous system (ANS) heartbeat modulation. While within-brain resting state networks have been the object of intense functional Magnetic Resonance Imaging (fMRI) research, technological and methodological limitations have hampered research on the central correlates of cardiovascular control dynamics. Here we combine the high temporal and spatial resolution as well as data volume afforded by the Human Connectome Project with a probabilistic model of heartbeat dynamics to characterize central correlates of sympathetic and parasympathetic ANS activity at rest. We demonstrate an involvement of a number of brain regions such as the Insular cortex, Frontal Gyrus, Lateral Occipital Cortex, Paracingulate and Cingulate Gyrus and Precuneous Cortex, as well as subcortical structures (Thalamus, Putamen, Pallidum, Brain-Stem, Hippocampus, Amygdala, and Right Caudate) in the modulation of ANS-mediated cardiovascular control, possibly indicating a broader definition of the central autonomic network (CAN). Our findings provide a basis for an informed neurobiological interpretation of the numerous studies which employ HRV-related measures as standalone biomarkers in health and disease.

Keywords: Neural signal processing, Neuromuscular systems - Computational modeling
Abstract: In this paper, an extension of the Hodgkin-Huxley model has been presented to describe the interaction between the nerve fibres. Based on the equivalent electrical circuit, the conductance per unit area between two coupled axons has been introduced as the pairwise coupling factors. Based on the measurement of the membrane potential, these coupling factors can be estimated by linearised identification methods. Therefore, the axon-to-axon interaction can be characterized, which improves the accuracy of the models. Comparing with the simulation studies of the mechanistic model, the presented model is more convenient to be analyzed while the computational complexity has been reduced. Using the presented description, the simulation results indicate that the presented model and the analysis strategy are effective.

Keywords: Neural signal processing
Abstract: Developing neuronal populations are assumed to increase their synaptic interactions and generate synchronized activity, such as bursting, during maturation. These effects may arise from increasing interactions of neuronal populations and increasing simultaneous intra-population activity in developing networks. In this paper, we investigated the neuronal network activity and its complexity by means of self-similarity during neuronal network development. We studied the phenomena using computational neuronal network models and actual in vitro microelectrode array data measured from a developing neuronal network of dissociated mouse cortical neurons. To achieve this, we assessed the spiking and bursting characteristics of the networks, and computed the signal complexity with Sample Entropy. The results show that we can relate increasing simultaneous activity in a neuronal population with decreasing entropy, and track the network development and maturation using this. We can conclude that the complexity of neuronal network signals decreases during the maturation. This can emerge from the fact that as networks mature, they exhibit more synchronous activity, thus decreasing the complexity of its signaling. However, increasing the number of interacting populations has lesser effect on the signal complexity. The entropy based measure provides a tool to assess the complexity of the neuronal network activity, and can be useful in the assessment of developing networks or the effects of drugs and toxins on their functioning.

Keywords: Neural signal processing, Human performance - Sensory-motor, Brain functional imaging - Source localization
Abstract: Sensorimotor control and the involvement of motor brain regions has been extensively studied, but the role nonmotor brain regions play during movements has been overlooked. This is particularly due to the difficulty of recording from multiple regions in the brain during motor control. In this study, we utilize stereoelectroencephalography (SEEG) recording techniques to explore the role nonmotor brain areas have on the way we move. Nine humans were implanted with SEEG depth electrodes for clinical purposes, which rendered access to local field potential (LFP) activity in deep and peripheral nonmotor structures. Participants performed fast and slow arm reaching movements using a robotic manipulandum. In this study, we explored whether neural activity in a given nonmotor brain structure correlated to movement path metrics including: path length, path deviation, and path speed. Statistical analysis revealed correlations between averaged neural activity in middle temporal gyrus, supramarginal gyrus, and fusiform gyrus and our path metrics both within and across the subjects. Furthermore, we split trials across subjects into two groups: one group consisted of trials with high values of each path metric and the other with low values. We then found significant differences in LFP power in specific frequency bands (e.g. beta) during movement between each group. These results suggest that nonmotor regions may dynamically encode path-related information during movement.

Keywords: Neural signal processing, Neural signals - Information theory
Abstract: Aim of this paper is proposing a stochastic model for vesicle release process, a part of neuro-spike communication. Hence, we study biological events occurring in this process and use microphysiological simulations to observe functionality of these events. Since the most important source of variability in vesicle release probability is opening of voltage dependent calcium channels (VDCCs) followed by influx of calcium ions through these channels, we propose a stochastic model for this event, while using a deterministic model for other variability sources. To capture the stochasticity of calcium influx to presynaptic neuron in our model, we study its statistics and find that it can be modeled by a distribution defined based on Normal and Logistic distributions.

Keywords: Neural signal processing, Brain functional imaging - EEG, Brain-computer/machine interface
Abstract: The relation between music and emotions has been investigated for decades. Most of the studies focused on short clips and were designed with specific tasks. This paper investigated the emotional states from electroencephalogram (EEG) activities during music appreciation. An emotion evoked experiment paradigm was designed during music appreciation. The EEG signals were recorded in 15 healthy adults during the entire process of music listening. The band power change (BPC) and higher order crossing (HOC) features were extracted from the EEG signals. A correlation-based feature analysis approach was proposed to find the most relevant features in time, frequency and channel space domains. From the results, this method achieved the average accuracy of 67.2% for the classification of high and low valence in the combination of BPC and HOC features. A deeper understanding of the brain emotional patterns could be helpful in building an intelligent and friendly affective application.

Keywords: Neural signal processing, Neural stimulation - Deep brain, Neurological disorders
Abstract: Deep brain stimulation (DBS) is a surgical treatment for Parkinson's disease (PD) but, despite clinical efficacy, the mechanisms of DBS still require investigation. Recent evidence suggests that the temporal pattern of the electrical pulses may be critical to the therapeutic merit of DBS and carefully-designed, non-regular patterns could ameliorate some of the motor symptoms in PD. It is unclear, though, how different stimulation patterns affect the neural activity in the basal ganglia and whether this is related to the pathophysiology of PD. In this study, a non-human primate was treated with DBS of the subthalamic nucleus while single-unit recordings were collected in the animal’s globus pallidus internus (GPi). Three stimulation patterns were applied (one regular, two non-regular) and the stimulation effects on the GPi spike trains were assessed via point process modeling. On a preliminary set of 23 GPi neurons, we show that regular DBS maximized the neuronal complexity, which is a measure of the amount of information that a single neuron can encode, and significantly increased the dependency of the neurons' spike trains on the background ensemble activity through an articulated balance of excitation and inhibition. Overall, regular DBS caused the largest modulation in the neurons’ spiking pattern and the largest increment in encoding capabilities. Both results may be relevant to the mechanisms of therapeutic DBS.

Keywords: General and theoretical informatics - Deep learning and big data to knowledge, Sensor Informatics - Multi-sensor data fusion
Abstract: Signal processing is one of the key points in brain computer interface (BCI) application. The common methods in BCI signal classification include canonical correlation analysis (CCA), support vector machine (SVM) and so on. However, because BCI signals are very complex and valid signals often come with confounded background noise, many current classification methods would lose meaningful information embedded in human EEGs. Otherwise, due to the huge inter-subject variability with respect to characteristics and patterns of BCI signals, there often exists large difference of classification accuracy among different subjects. Since BCI signals have high dimensionality and multi-channel properties, this paper proposes a novel structure of deep belief neural (DBN) network stacked by restricted boltsman machine (RBM) to extract efficient features from steady-state motion visual evoked potential signals and implement further classification. Here DBN extracts local feature from BCI data of each channel separately and fuses the local features, and then input the fused features to the output classifier which is consist of softmax units. Results proved that the proposed algorithm could achieve higher accuracy and lower inter-subject variability in short response time when compared to conventional CCA method.

Keywords: General and theoretical informatics - Big data analytics, General and theoretical informatics - Causality analysis and case-based reasoning, Bioinformatics - Bioinformatics databases
Abstract: Excessive weight is connected with an increased risk of certain life-threatening diseases. However, some evidence shows that among patients with chronic diseases such as heart failure (HF) chronic kidney disease (CKD) and COPD, increased weight is paradoxically associated with a decreased risk of mortality. This counterintuitive phenomenon is referred to as the obesity paradox. The obesity paradox has been mostly observed among certain cohorts of patients with HF, but not specific to patients in the Intensive Care Unit (ICU) setting. This paper studies the relationship between obesity and mortality of ICU patients with and without HF and presents evidence supporting the existence of this paradox. The results provide helpful insights for developing more patient-centric care in ICUs. Additionally, we use both the MIMIC-II and (recently available) MIMIC-III databases, for which few comparative studies exist to date. We demonstrate an aspect of consistency between the databases, providing a significant step towards validating the use of the newly announced MIMIC-III in broader studies.

Keywords: General and theoretical informatics - Deep learning and big data to knowledge, General and theoretical informatics - Machine learning, General and theoretical informatics - Decision support systems
Abstract: This work investigated the potential of Deep Neural Network in detection of cerebral embolic signal (ES) from transcranial Doppler ultrasound (TCD). The resulting system is aimed to couple with TCD devices in diagnosing a risk of stroke in real-time with high accuracy. The Adaptive Gain Control (AGC) approach developed in our previous study is employed to capture suspected ESs in real-time. By using spectrograms of the same TCD signal dataset as that of our previous work as inputs and the same experimental setup, Deep Convolutional Neural Network (CNN), which can learn features while training, was investigated for its ability to bypass the traditional handcrafted feature extraction and selection process. Extracted feature vectors from the suspected ESs are later determined whether they are of an ES, artifact (AF) or normal (NR) interval. The effectiveness of the developed system was evaluated over 19 subjects going under procedures generating emboli. The CNN-based system could achieve in average of 83.0% sensitivity, 80.1% specificity, and 81.4% accuracy, with considerably much less time consumption in development. The certainly growing set of training samples and computational resources will contribute to high performance. Besides having potential use in various clinical ES monitoring settings, continuation of this promising study will benefit developments of wearable applications by leveraging learnable features to serve demographic differentials.

Keywords: General and theoretical informatics - Deep learning and big data to knowledge, Health Informatics - Patient tracking, Health Informatics - Informatics for chronic disease management
Abstract: There are extensive studies investigating congestive heart failure (CHF) detection based on heart rate variability. Although a high level of accuracy has been achieved, its robustness under different conditions is not guaranteed. To improve the robustness, we applied sparse auto-encoder-based deep learning algorithm in CHF detection with RR intervals. A total data size of 30,592 (5-min RR interval) was obtained from 72 healthy persons and 44 CHF patients. The deep learning algorithm first extracts unsupervised features using a sparse auto-encoder from raw RR intervals, then constructs a deep neural network model with various hidden nodes combinations. Results showed that the model achieved 72.41% accuracy. This demonstrated that RR intervals have potential in CHF detection but cannot fully reflect dynamic change in 24-h.

Keywords: General and theoretical informatics - Deep learning and big data to knowledge, Health Informatics - Decision support methods and systems, General and theoretical informatics - Big data analytics
Abstract: Multiple cause of death data provides a valuable source of information that can be used to enhance health standards by predicting health related trajectories in societies with large populations. These data are often available in large quantities across U.S. states and require Big Data techniques to uncover complex patterns that are hidden in them. We design two different classes of models suitable for large-scale analysis of mortality data, a Hadoop based ensemble of random forests trained over N-grams, and the DeepDeath, a deep classifier based on recurrent neural networks paradigm. We apply both classes to the mortality data provided by the National Center for Health Statistics and show that while both perform significantly better than the random classifier, the deep model that utilizes long short-term memory networks (LSTMs), surpasses the N-gram based models and is capable of learning the temporal aspect of the data without a need for building ad-hoc, expert-driven features.

Keywords: General and theoretical informatics - Deep learning and big data to knowledge, General and theoretical informatics - Artificial Intelligence, Sensor Informatics - Wireless sensors and systems
Abstract: Levodopa is the gold standard therapy for Parkinson’s disease (PD), but its prolonged usage leads to additional motor complications, namely levodopa-induced dyskinesia (LID). To assess LID and adjust drug regimens for optimal relief, patients attend regular clinic visits. However, the intermittent nature of these visits can fail to capture important changes in a person’s condition. With the recent emergence of deep learning achieving impressive results in a wide array of fields including computer vision, there is an opportunity for video analysis to be used for automated assessment of LID. Deep learning for pose estimation was studied as a viable means of extracting body movements from PD assessment videos. Movement features were computed from joint trajectories. Results show that features derived from vision-based analysis have moderate to good correlation with clinician ratings of dyskinesia severity. This study presents the first application of deep learning to video analysis in PD, and demonstrates promise for future development of a non-contact system for objective PD assessment.

Keywords: Cardiac catheterization, Cardiovascular and respiratory system modeling - Blood flow models
Abstract: A decent guidewire behavior simulation is vital to the virtual vascular intervention training. The influence of blood flow has rarely been taken into consideration in former works of guidewire simulation. This paper addresses the problem by integrating blood flow analysis and proposes a novel guidewire simulation model. The blood flow distribution inside arterial vasculature is computed by separating the vascular model into discrete cylindrical vessels and modeling the flow in each vessel with the Poiseuille Law. The blood flow computation is then integrated into a Kirchhoff rods model. The simulation could be run in real time with hardware acceleration at least 30 fps. To validate the result, an experiment environment with a 3D printed vascular phantom and an electromagnetic tracking(EMT) system was set up with clinical-used guidewire sensors applied in phantom to trace its motion as the standard for comparison. Experiment results reveal that the shown blood flow effected model presents better physical credibility with a lower and more stable root-mean-square(RMS) at 2.14mm±1.24mm, better than the Kirchhoff model of 4.81mm±3.80mm.

Keywords: Vascular mechanics and hemodynamics - Vascular Disease, Vascular mechanics and hemodynamics - Vascular Hemodynamics
Abstract: Although flow-diversion (FD) treatment has been proven to be able to induce intracranial aneurysm (IA) occlusion, clinical follow-ups reported that a number of patients may still suffer from delayed IA rupture or incomplete aneurysm occlusion post-treatment. Complete aneurysm occlusion is believed to be associated with favourable haemodynamic alteration post-treatment, which may be greatly affected by the selection of device size and quantity, as well as the FD deployment procedure. However, clinicians have to choose and deploy the FD relying on their experience, since no post-stenting haemodynamic information is generally available to them prior to a specific treatment. In this study, using a virtual FD deployment technique and computational fluid dynamics method, we demonstrate and compare the haemodynamic changes after virtual FD treatments using a variety of prospective treating strategies.

Keywords: Vascular mechanics and hemodynamics - Vascular Disease, Vascular mechanics and hemodynamics - Vascular Hemodynamics
Abstract: The flow-diverting (FD) stent has become a commonly used endovascular device to treat cerebral aneurysms. This discourages blood from entering the aneurysm, thereby reducing the likelihood of aneurysm rupture. Using computational fluid dynamics (CFD) to simulate the aneurysmal haemodynamics after FD treatment could help clinicians predict the stent effectiveness prior to the real procedure in the patient. As an alternative to modelling the stent as a fine wire mesh, modelling the FD stent as a porous medium was established to save computational time, and has also been proved capable of predicting the same haemodynamics as obtained using the real FD stent geometry. The flow resistance effect of a porous-medium stent may differ with respect to its morphology or permeability; however, the flow resistance effect after adjusting these parameters had not been clarified. In this study, we analysed the haemodynamic changes caused by alterations of porous-medium thickness and permeability, thereby providing future porous-medium stent simulations with important information on the respective parametric sensitivities. We found significant sensitivity to permeability. Results were insensitive to thickness when permeability was adjusted beforehand to compensate. We also compared our results with observations from an in-vitro model, and found good agreement. This supports adoption of porous-medium models in future work.

Keywords: Cardiac catheterization, Cardiovascular and respiratory system modeling - Cardiovascular Disease, Coronary artery disease
Abstract: Robotic surgical systems are becoming increasingly popular for the treatment of cardiovascular diseases. However, most of them have been designed without considering techniques and skills of natural surgical manipulations, which are key factors to clinical success of percutaneous coronary intervention. This paper proposes an HMM-based framework to recognize six typical endovascular manipulations for surgical skill analysis. A simulative surgical platform is built for endovascular manipulations assessed by five subjects (1 expert and 4 novices). The performances of the proposed framework are evaluated by three experimental schemes with the optimal model parameters. The results show that endovascular manipulations are recognized with high accuracy and reliable performance. Furthermore, the acceptable results can also be applied to the design of next generation vascular interventional robots.

Keywords: Vascular mechanics and hemodynamics - Vascular Hemodynamics, Vascular mechanics and hemodynamics - Vascular Disease, Vascular mechanics and hemodynamics - Vascular mechanics
Abstract: Coil embolization using a micro catheter and coils is one of the most popular surgical methods used for treating intracranial aneurysms. This method is minimally invasive, however, recanalization remains a major problem. To prevent aneurysmal recanalization, surgeons need to better understand the effects of changing catheter position because this is under their control during operations. However, our latest study only considered a sidewall-type aneurysm and the tip of the catheter was always placed at the center of the aneurysm. There are no other studies that use FEM and CFD to investigate the ideal catheter position for effective coil embolization. In this study, we use the finite element method and computational fluid dynamics to simulate coil embolization. A basic bifurcation-type aneurysm model was created and the embolic coil was modeled based on actual Stryker Target 360 Soft Coils. We evaluated the reduction in velocity after embolization and the distance of the embolized coil’s center of gravity from the center of the aneurysm while changing the catheter tip position. The results of this study indicate that the formations of the embolized coil took multifarious shapes when the catheter position was changed. Although the flows are reduced after the embolism in all cases, the percentage of velocity reduction completely differs for each case. We also observed a relationship between the velocity reduction and neck volume embolization ratio (NVER) with a correlation coefficient of 0.912. In addition, the relations between the Z-coordinate of the catheter tip and rCG shows a linear regression with a correlation coefficient of 0.750. The neck volume embolization ratio is also important for effective velocity reduction in both bifurcation-type and sidewall-type aneurysms. In addition, it is useful to set the catheter tip at the bottom of the aneurysms to deploy the first coil as a flaming coil.

Keywords: Cardiovascular and respiratory system modeling - Cardiovascular-Respiratory Interactions, Cardiovascular and respiratory system modeling - Cardiac models, Cardiovascular and respiratory system modeling - Compartmental modeling
Abstract: Mechanical heart-lung interactions are often overlooked in clinical settings. However, their impact on cardiac function can be quite significant. Mechanistic physiology-based models can provide invaluable insights into such cardiorespiratory interactions, which occur not only under external mechanical ventilatory support but in normal physiology as well. In this work, we focus on the cardiac component of a previously developed mathematical model of the human cardiopulmonary system, aiming to improve the model's response to the intrathoracic pressure variations that are associated with the respiratory cycle. Interventricular septum and pericardial membrane are integrated into the existing model. Their effect on the overall cardiac response is explained by means of comparison against simulation results from the original model as well as experimental data from literature.

Keywords: Image segmentation, X-ray CT imaging
Abstract: The conventional graph cuts technique has been widely used for image segmentation due to its ability to find the global minimum and its ease of implementation. However, it is an intensity-based technique and as a result is limited to segmentation applications where there is significant contrast between the object and the background. We modified the conventional graph cuts method by adding shape prior and motion information. Active shape models (ASM) with signed distance functions were used to capture the shape prior information, preventing unwanted surrounding tissue from becoming part of the segmented object. The optical flow method was used to estimate the local motion and to extend 3D segmentation to 4D by warping a prior shape model through time. The method has been applied to segmentation of whole lung boundary and whole liver boundary from respiratory gated CT data. 4D lung boundary segmentation in five patients, and 4D liver boundary segmentation in five patients were performed and in each case, results were compared with the results from expert-delineated ground truth. 4D segmentation for five phases of CT data took approximately ten minutes on a PC workstation with AMD Phenom II and 32GB of memory. An important by-product is quantitative whole organ volumes from respiratory gated CT from end-inspiration to end-expiration which can be determined with high accuracy.

Keywords: Image segmentation, Multivariate image analysis, Functional image analysis
Abstract: RADEval is a tool developed to assess the expected clinical impact of contouring accuracy when comparing manual contouring and semi-automated segmentation. The RADEval tool, designed to process large scale datasets, imported a total of 2,760 segmentation datasets, along with a Simultaneous Truth and Performance Level Estimation (STAPLE) to act as ground truth tumor segmentations. Virtual dose-maps were created within RADEval and two different tumor control probability (TCP) values using a Logistic and a Poisson TCP models were calculated in RADEval using each STAPLE and each dose-map. RADEval also virtually generated a ring of normal tissue. To evaluate clinical impact, two different uncomplicated TCP (UTCP) values were calculated in RADEval by using two TCP-NTCP correlation parameters (δ= 0 and 1). NTCP values showed that semi-automatic segmentation resulted in lower NTCP with an average 1.5 – 1.6 % regardless of STAPLE design. This was true even though each normal tissue was created from each STAPLE (p < 0.00001). TCP and UTCP presented no statistically significant differences (p ≥ 0.1884). The intra-operator standard deviations (SDs) for TCP, NTCP and UTCP were significantly lower for the semi-automatic segmentation method regardless of STAPLE design (p < 0.0331). Both intra-and inter-operator SDs of TCP, NTCP and UTCP were significantly lower for semi-automatic segmentation for the STAPLE 1 design (p <0.0331). RADEval was able to efficiently process 4,920 datasets of two STAPLE designs and successfully assess the expected clinical impact of contouring accuracy.

Keywords: Image segmentation
Abstract: Tooth contour segmentation from dental computerized tomography (CT) images is one of the fundamental steps in reconstructing the three-dimensional models of teeth. However, existing methods depend on the tooth shape similarity between adjacent slices, and have difficulty to segment these angled teeth whose contour shapes from adjacent slices may differ a lot. This study proposes a new method for contour segmentation of angled teeth from CT images. The volume of interest (VOI) of target tooth and corresponding tooth axis are first extracted from volumetric CT images. Local images within the VOI of target tooth are then rotated such that the tooth axis in the rotated images is perpendicular to the transverse section. Tooth contours are finally segmented from the rotated images using a hybrid level set model slice-by-slice. Experimental results verified that the proposed method was effective to segment contours of angled teeth from CT images.

Keywords: Image segmentation
Abstract: The Nuchal translucency (NT), which is the collection of fluid at the back of the fetal neck, is related to chromosomal defects and early cardiac failure in first trimester of pregnancy. In clinic, the thickness of NT is used as an important marker in prenatal screening, and is manually measured by sonographers in the mid-sagittal plane. In this paper, an automatic method based on dynamic programming is proposed to detect the thickness and area of NT in the mid-sagittal plane. Furthermore, the volume of NT in the whole three-dimensional ultrasound data is also measured. A novel cost function for dynamic programming is proposed and results in higher accuracy of NT border detection. As the nuchal translucency is the collection fluid part, higher dimensional markers of NT possess more potential to represent the amount of the fluid.

Keywords: Image segmentation, Brain image analysis, Image feature extraction
Abstract: Skull stripping, which refers to the segmentation of brain tissue from non-brain tissue, has been challenging due to the ramification of the human brain structures and volatile parameters in the magnetic resonance imaging (MRI) procedures. It has been one of the most critical preprocessing steps in medical image analysis. We propose a hybrid skull stripping algorithm that is based on texture feature analysis, fuzzy possibilistic c-means (FPCM), and morphological operations. The input MR image is first processed to obtain two texture feature maps, to which the FPCM is applied for acquiring brain and non-brain masks. A number of morphological operations are subsequently performed to extract the brain. Our algorithm has been compared with two famous methods and evaluated on the internet brain segmentation repository (IBSR) datasets. Preliminary experimental results suggested that this new framework achieved high accuracy and outperformed the compared methods. We believe that the proposed scheme is of effectively potential in a wide variety of brain MR image segmentation applications.

Keywords: Image segmentation, Image feature extraction, Brain image analysis
Abstract: Irreversible infarcts are critical for the assessment of potential risk and benefit pertaining to thrombolysis in hyper-acute ischemic stroke. It is a challenging work to segment infarct at hyper-acute stage due to the substantial variability. A general abnormal tissue segmentation method is proposed and applied to segment hyper-acute ischemic infarct in this paper. Multiple features are designed to train a random forest classifier for voxel classification. Sparse coding based bag-of-features is adopted to train a region classifier for infarct region recognition. The proposed method has been validated on 98 consecutive patients recruited within 6 hours from onset and achieved a higher Dice coefficient 0.774±0.117 than the other two existing methods (0.755±0.118; 0.597±0.204). It could provide a potential tool to quantify infarcts from diffusion weighted imaging at hyper-acute stage with accuracy to assist the decision making especially for thrombolytic therapy.

Keywords: Point of care - Home-based applications, Point of care - Technologies in resource limited settings
Abstract: Burn injury is a major public health issue in developing countries, with most injuries being largely associated with the use of kitchen stoves. This study details the development of a cost-effective gesture and voice recognition controlled (GVC) system to be used by individuals with disabilities to reduce the likelihood of burn injury and improve their quality of life. The device replaces conventional dial controls with voice and hand gesture recognition sensors and software which are designed to be easily implemented into a household kitchen. Preliminary evaluation of the GVC system’s performance in gesture and voice recognition, gas leak detection and ignition control tests were conducted using a Bunsen burner as a stove top surrogate. The voice and gesture recognition tests yielded sensitivities of 88% and 100%, respectively. These results suggest that the GVC system may be a promising solution for burn injury prevention pending further work to improve its reliability and robustness.

Keywords: Point of care - Respiratory monitoring, Point of care - Detection and monitoring, Medical technology - Design and development
Abstract: Capnometry is a method to measure carbon dioxide (CO2) in exhaled gas and it has been used in patients to monitor their respiratory status. Monitoring of exhaled CO2 during endoscopic procedures has been shown to be effective in detecting drug-induced respiratory depression. Oxygen (O2) supplementation is given to patients to abolish hypoxia during endoscopy. However, oxygen administration can interfere with CO2 measurement owing to oxygen flow. Therefore, we developed cap-ONE Biteblock for patients undergoing endoscopy with oxygen supply to measure CO2 accurately. In this study we evaluated the basic performance of cap-ONE Biteblock. The cap-ONE Biteblock system could accurately measure CO2 and efficiently supply O2 comparing to conventional devices via a bench study.

Keywords: Point of care - Home-based applications, Information communication and networking - mHealth innovations, Point of care - Heart rate monitoring
Abstract: In this paper, an implementation of mobile-Visible Light Communication (mVLC) technology for clinical data transmission in home-based mobile-health (mHealth) rehabilitation system is introduced. Mobile remote rehabilitation program is the solutions for improving the quality of care of the clinicians to the patients with chronic condition and disabilities. Typically, the program inquires routine exercises which obligate patients to wear wearable electronic sensors for hours in a specific range of time. Thus it motivate us to develop a novel harmless biomedical communicating system since most of the device's protocol was based on RF communication technology which risky for a human body in term of long term usage due to RF exposure and electromagnetic interference (EMI). The proposed system are designed to utilize a visible light as a medium for hazardless-communication between wearable sensors and a mobile interface device (smartphone). Multiple clinical data such as photoplethysmogram (PPG), electrocardiogram (ECG), and respiration signal are transmitted through LED and received by a smartphone camera. Furthermore, a smartphone also used for local interface and data analyzer henceforth sent the data to the cloud for further clinician’s supervision.

Keywords: Point of care - Technologies for slightly trained operators, Point of care - Technologies in resource limited settings, Point of care - Detection and monitoring
Abstract: This work focuses on the development of a sophisticated technique via STR typing to unequivocally verify the authenticity of urine samples before sent to laboratories. STR profiling was conducted with the CSF1PO, TPOX, TH01 Multiplex System coupled with a smartphone-based detection method. The promising capability of the method to identify distinct STR profiles from urine of different persons opens the possibility to conduct sample authenticity tests. On-site STR profiling could be realized with a self-contained autonomous device with an integrated PCR microchip shown hereby.

Keywords: Point of care - Detection and monitoring, Point of care - Home-based applications, Global healthcare challenges
Abstract: Falls are a major risk for elderly people’s health and independence. Fast and reliable fall detection can improve chances of surviving the accident and coping with its physical and psychological consequences. Recent research has come up with various solutions, all suffering from significant drawbacks, one of them being the intrusiveness into patient’s life. This paper proposes a novel fall detection system based on a sensitive floor sensor made out of a piezoelectric material and a machine learning approach. The detection is done by a combination between a supervised Random Forest and an aggregation of its output over time. The database was made using acquisitions from 28 volunteers simulating falls and other behaviours. Unlike existent fall detection systems, our solution offers the advantages of having a passive sensor (no power supply is needed) and being completely unobtrusive since the sensor comes with the floor. Results are compared with state-of-the-art classification algorithms. On our database, good performance of fall detection was obtained with a True Positive Rate of 94.4% and a False Positive Rate of 2.4%.

Keywords: Point of care - Respiratory monitoring, Point of care - Heart rate monitoring
Abstract: This paper addresses non-contact monitoring of physiological signals induced by respiration and heartbeat. To detect the tiny physiological movements of the chest or other parts of the torso, a Mulitple-Input Multiple-Output (MIMO) radar is used. The spatially distributed transmitters and receivers are able to detect the chest surface movements of one or multiple persons in a room. Due to several bistatic measurements at the same time a robust detection and measuring of the breathing and heartbeat rate is possible. Using an appropriate geometrical configuration of the sensors even a localization of the person is feasible.

Keywords: Image visualization
Abstract: Motor-skill learning for complex robotic tasks is challenging due to high task variability. In this study, we propose a user-friendly tool for Learning from Demonstration (LfD) that relies on the use of Bayesian nonparametric dimensionality reduction. We implement our framework in an assistive robotics setting for imparting motor-skills to a dual-arm robot performing clothing assistance tasks.

Keywords: Image visualization
Abstract: This study aimed to decrease risk of falls via displacement of the axilla during crutch walking. We developed an axillary pressure feedback system that provided visual and/or auditory information to crutch users. Results suggested that this feedback system was effective in beginner crutch users.

Keywords: Image classification, Image feature extraction
Abstract: In this paper, we introduce an ensemble classification method using local images for diagnosis support system based on intraoral images. In the developing system, dentists are required to manually cutout a local image including disease area from the intraoral image. This manual operation includes serious individual difference, and it causes decrease of accuracy. By applying a concept of ensemble method, influence of individual difference can be weakened.

Keywords: Optical imaging
Abstract: Ocular vessels are an important indicator in ophthalmology, but it is difficult to observe them at a glance because the ocular blood vessels are very small and the structure is complex. In this study, we propose a new method of analyzing the ocular vasculature using photoacoustic micriscopy and random sample consensus algorithm. We found the eyeball surface and reconstructed an image that is easier to analze the ocular vasculature than the conventional visualization method.

Keywords: Signal pattern classification, Physiological systems modeling - Signal processing in simulation, Principal and independent component analysis - Blind source separation
Abstract: We describe the extension of pre-existing cylindrical volume conductor model to synthetic high-density surface electromyograms (hdEMG), simulated during dynamic contractions of fusiform skeletal muscles. Its modular structure comprises two main parts. First, dynamic changes of motor unit action potentials (MUAPs) during 36 discrete steps of muscle shortening are simulated. Second, the increase in depth of simulated motor units (MUs) due to shortening and thickening of muscle fibers is simulated. MU firing patterns are generated with the model proposed by Fuglevand et al. and convolved with simulated MUAPs. In this way, the hdEMG simulator can be used to generate dynamic hdEMG of arbitrary muscle shortening, thickening and excitation profiles. In order to demonstrate the value of the aforementioned simulator we independently analyze the impact of muscle shortening and muscle thickening on MU identification by Convolution Kernel Compensation (CKC) technique.

Keywords: Time-frequency and time-scale analysis - Empirical mode decomposition in biosignal analysis, Nonlinear dynamic analysis - Nonlinear filtering
Abstract: The use of the Empirical Mode Decomposition (EMD) for nonlinear signal processing has been popularized in the recent years. However, its utility for the processing of multichannel Electromyography (EMG) signals is still limited. This paper investigates the decomposition performance of multichannel EMGs by using EMD-based approaches, Ensemble EMD (EEMD), Multivariate EMD (MEMD), and Noise-Assisted MEMD (NA-MEMD). In the experiment, 11 male subjects undergo three exercise programs, leg extension from a sitting position, flexion of the leg up, and gait, while electrodes are placed on the muscle groups, biceps femoris, vastus medialis, rectus femoris, and semitendinosus. The outcomes are then quantitatively estimated on the basis of three criterions, the number of Intrinsic Mode Functions (IMFs), mode-alignment and mode-mixing. Results show both MEMD and NA-MEMD can guarantee equal numbers of IMFs, whereas for mode-alignment and mode-mixing, NA-MEMD is optimal compared with MEMD and EEMD, and MEMD is merely better than EEMD. This finding implies that NA-MEMD is effective for simultaneously analyzing IMFs based frequency bands. It has a vital clinical implication in exploring the neuromuscular patterns that enable the multiple muscle groups to coordinate while performing functional activities of daily living.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signals and systems
Abstract: In this study we investigated a technique for estimating the progression of localized muscle fatigue. This technique measures the dependence between motor units using high density surface electromyogram (HD-sEMG) and is based on the Normalized Mutual Information (NMI) measure. The NMI between every pair combination of the electrode array is computed to measure the interactions between electrodes. Participants in the experiment had an array of 64 electrodes (16 by 4) placed over the TA of their dominate leg such that the columns of the array ran parallel with the muscle fibers. The HD-sEMG was recorded whilst the participants maintained an isometric dorsiflexion with their dominate foot until task failure at 40% and 80% of their maximum voluntary contraction (MVC). The interactions between different locations over the muscle were computed using the recorded HD-sEMG signals. The results show that the average interactions between various locations over the TA significantly increased during fatigue at both levels of contraction. This can be attributed to the dependence in the motor units.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Connectivity measurements, Data mining and processing in biosignals
Abstract: Exercises under isometric and dynamic contractions are influenced by the rate coding and recruitment strategies. The study of muscle strength under dynamic contraction is normally performed using one-repetition maximum (1-RM) method. There are several variants of deriving one repetition method using number of repetitions and load that are useful in physical fitness and clinical rehabilitation program. However, the factors of dynamic contractions such as endurance time, speed of muscle contractions and muscle activity are not considered in 1-RM methods. The muscular activities are analyzed using surface electromyography (sEMG) signals. Limited work has been reported on the relationship between the 1-RM method and factors such as endurance time, speed of contraction and sEMG activity. In this work, a modified 1-RM method is proposed, namely, N-RM, using load, number of repetitions, endurance time, speed of contraction and normalized sEMG activity. For this purpose, sEMG signals are recorded from 58 healthy subjects under standard dynamic contraction protocol involving curl exercise. Conventional 1-RM is computed by using Epley’s method and compared with proposed method using correlation analysis. The results show that 1-RM increases linearly with number of curls (r=1) but has a poor correlation coefficient with sEMG (r=0.01) and endurance time (r =0.4). The curl speed for lower 1-RM and higher 1-RM did not show any statistical difference (p =0.2). The proposed N-RM is observed to have good correlation with endurance time (r=0.734), curl speed (r=0.893) and sEMG activity (r=0.8851). These results demonstrate that the proposed N-RM is highly correlated to factors influencing the dynamic contractions. This method can be further extended to assess muscles under various clinical disorders and sports training.

Keywords: Physiological systems modeling - Signal processing in physiological systems
Abstract: Analysis of muscle co-contractions seems to be relevant in the characterization of children pathologies such as spastic cerebral palsy. The aim of the study was the quantification of thigh-muscle co-contractions during walking in healthy children. To this aim, the Statistical Gait Analysis, a recent methodology providing a statistical characterization of gait, was performed on surface EMG signals from Vastus Medialis (VM) and Lateral Hamstrings (LH) in 30 healthy 6-to-8-year-old children. Muscular co-contraction was assessed as the overlapping period between activation intervals of agonist and antagonist muscles. As in adults, VM activity occurring from terminal swing to the following loading response superimposed LH activity in the same percentage of the gait cycle. This co-contraction occurred in order to control knee joint stability during weight acceptance. It was acknowledged in the totality (100 %) of the considered strides. Concomitant activity of VM and LH was detected also in the second half of stance phase in 17.1 ± 4.8 % of the considered strides. Working VM and LH on different joints, this concomitant activity of antagonist muscles should not be considered as an actual co-contraction. Present findings provide new information on the variability of the reciprocal role of VM and LH during child walking, useful for comparison between normal and pathological walking in the clinical context and for designing future studies on maturation of gait.

Keywords: Diagnostic devices - Physiological monitoring, Clinical engineering, Ambulatory Diagnostic devices - Point of care technologies
Abstract: Involuntary muscle activities like fasciculations or tremor are an indication for several neurological disorders. However, currently used techniques for measuring those activities are limited due to their invasiveness, the unsuitability for measuring a whole body simultaneously and the lack of an objective measurement of amplitude and duration of muscle activity. Hence, we developed a new laser-based sensor for the remote quantification of muscle activity. In the present paper we show a basic evaluation of our system by reference to ultrasound measurements. Our results show the detection limits of our remote sensor technology in terms of fasciculation size and depth within the muscle. Those results will help us for a better interpretation of our measurement results and hold promise for the future development of our system.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signal processing in physiological systems, Signal pattern classification
Abstract: Emphatic doctor-patient communication has been associated with an improved psycho-physiological well-being involving cardiovascular and neuroendocrine responses. Nevertheless, a comprehensive assessment of heartbeat linear and nonlinear/complex dynamics throughout the communication of a life-threatening disease has not been performed yet. To this extent, we here study heart rate variability (HRV) series gathered from 17 subjects while watching a video where an oncologist discloses the diagnosis of a cancer metastasis to a patient. Further 17 subjects watched the same video including additional affective emphatic contents. For the assessment of the two groups, linear heartbeat dynamics was quantified through measures defined in the time and frequency domains, whereas nonlinear/complex dynamics referred to measures of entropy, and combined Lagged Poincare Plots (LPP) and symbolic analyses. Considering differences between the beginning and the end of the video, results from non-parametric statistical tests demonstrated that the group watching emphatic contents showed HRV changes in the LF/HF ratio exclusively. Conversely, the group watching the purely informative video showed changes in vagal activity (i.e., HF power), LF/HF ratio, as well as LPP measures. Additionally, a Support Vector Machine algorithm including HRV nonlinear/complex information was able to automatically discern between groups with an accuracy of 76.47%. We therefore propose the use of heartbeat nonlinear/ complex dynamics to objectively assess the empathy level of healthy women.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signal processing in physiological systems, Time-frequency and time-scale analysis - Nonstationary processing
Abstract: The heart-rate fractal dynamics can be assessed by Detrended Fluctuation Analysis (DFA), originally proposed for estimating a short-term coefficient, alpha1 (for scales n≤12 beats), and a long-term coefficient alpha2 (for longer scales). Successively, DFA was extended to provide a multiscale alpha, i.e. a continuous function of n, alpha(n); or a multifractal alpha, i.e. a function of the order q of the fluctuations moment, alpha(q). Very recently, a multifractal-multiscale DFA was proposed for evaluating multifractality at different scales separately. Aim of this work is to describe the multifractal multiscale dynamics of three cardiovascular signals often recorded beat by beat in physiological and clinical settings: systolic blood pressure (SBP), diastolic blood pressure (DBP) and pulse interval (PI, inverse of the heart rate). We recorded SBP, DBP and PI for at least 90’ in 65 healthy volunteers at rest, and adapted the previously proposed multifractal multiscale DFA to estimate alpha as function of the temporal scale, tau, between 15 and 450 s, and of the order q, between -5 and 5. We report, for the first time: 1) substantial differences among alpha(q,tau) surfaces of PI, SBP and DBP; 2) a strong dependency of the degree of multifractality on the temporal scale.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Multivariate signal processing
Abstract: The aim of this study is to characterize cardiovascular and respiratory signals during orthostatic and mental stress as reflected in indices of entropy and complexity, providing a comparison between the performance of different estimators. To this end, the heart rate variability, systolic blood pressure, diastolic blood pressure and respiration time series were extracted from the recordings of 61 healthy volunteers undergoing a protocol consisting of supine rest, head-up tilt test and mental arithmetic task. The analysis was performed in the information domain using measures of entropy and conditional entropy, estimated through model-based (linear) and model-free (binning, nearest neighbor) approaches. Our results show that different types of stress elicited different responses in the employed indices. On one hand, entropy mainly reflected known changes in the variance of physiological time series. On the other hand, the information conveyed by conditional entropy allowed to characterize the complexity of the four time series during the two stress tasks: we found that cardiac and vascular dynamics underwent a reduction in complexity as a consequence of postural stress, while vascular and respiratory complexity increased as a result of mental stress. As for the performance of different estimators, we did not find substantial differences between model-based and model-free approaches, possibly indicating that significant non-linear dynamics did not appear in the studied conditions.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Signal pattern classification
Abstract: The most recently introduced concept of a ‘complete entropy profile’ is a non-parametric (with regard to tolerance r) approach of entropy estimation. Given a signal, on generating its complete entropy profile, numerous secondary measures of regularity can be derived from the same. These profile based measures are seen to outperform the traditional ApEn statistic (evaluated at a single r) in estimating signal regularity. In this paper, we compare the performance of ApEn (evaluated at an r=0.15*SD of signal and an m=2) with that of profile based measures such as MaxApEn, TotalApEn, AvgApEn, SDApEn, kurtApEn and skewApEn, in detecting ‘Arrhythmic’ RR interval signals from ‘Normal’ RR interval signals. Results indisputably prove the superiority of AvgApEn (AUC>0.9 at data lengths Ngeq200) and MaxApEn (AUC>0.75 at all data lengths) as regularity statistics in detecting Arrhythmia, above all the other measures used.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Multivariate signal processing, Physiological systems modeling - Signal processing in physiological systems
Abstract: Abstract— In this work, a nonlinear method to study multivariate interactions, called multivariate symbolic dynamics (MSD), was introduced. The usefulness of this technique was studied on respiratory-cardiovascular data from young women with vasovagal syncope (VVS) and from healthy subjects. The study included 16 female patients diagnosed with VVS and 24 age-matched healthy subjects (12 women). All subjects were enrolled in a head-up tilt (HUT) test, breathing normally, including 5 min of supine position and 18 to 28 min of 70° orthostatic phase. The MSD parameters were dynamically obtained for 5-min windows shifted by 1 min during HUT test. In supine position there were no considerable differences. During orthostatic phase, parameters from MSD showed a highly significantly (p=0.00005) increased occurrence of impaired respiratory-cardiovascular interactions in female patients susceptible to vasovagal syncope. This study provided promising results for a new multivariate method to investigate respiratory-cardiovascular interactions.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification
Abstract: In this work, the cardiorespiratory pattern of patients undergoing extubation process is studied. First, the respiratory and cardiac signals were resampled, next the Symbolic Dynamics (SD) technique was implemented, followed of a dimensionality reduction applying Forward Selection (FS) and Moving Window with Variance Analysis (MWVA) methods. Finally, the Linear Discriminant Analysis (LDA) and Support Vector Machines (SVM) classifiers were used. The study analyzed 153 patients undergoing weaning process, classified into 3 groups: Successful Group (SG: 94 patients), Failed Group (FG: 39 patients), and patients who had been successful during the extubation and had to be reintubated before 48 hours, Reintubated Group (RG: 21 patients). According to the results, the best classification present an accuracy higher than 88.98 ± 0.013% in all proposed combinations.

Keywords: Signal pattern classification, Data mining and processing in biosignals, Physiological systems modeling - Multivariate signal processing
Abstract: Biometric technologies offer great advantages over other recognition methods, but there are concerns that they may compromise the privacy of individuals. In this paper, an electrocardiogram (ECG)-based cancelable biometric scheme is proposed to relieve such concerns. In this scheme, distinct biometric templates for a given beat bundle are constructed via "subspace collapsing." To determine the identity of any unknown beat bundle, the multiple signal classification (MUSIC) algorithm, incorporating a "suppression and poll" strategy, is adopted. Unlike the existing cancelable biometric schemes, knowledge of the distortion transform is not required for recognition. Experiments with real ECGs from 285 subjects are presented to illustrate the efficacy of the proposed scheme. The best recognition rate of 97.58 % was achieved under the test condition Ntrain = 10 and Ntest = 10.

Keywords: Data mining and processing in biosignals, Independent component analysis
Abstract: An adaptive Fourier decomposition (AFD) based R-peak detection method is proposed for noisy ECG signals. Although lots of QRS detection methods have been proposed in literature, most detection methods require high signal quality. The proposed method extracts the R waves from the energy domain using the AFD and determines the R-peak locations based on the key decomposition parameters, achieving the denoising and the R-peak detection at the same time. Validated by clinical ECG signals in the MIT-BIH Arrhythmia Database, the proposed method shows better performance than the Pan-Tompkin (PT) algorithm in both situations of a native PT and the PT with a denoising process.

Keywords: Data mining and processing - Pattern recognition, Data mining and processing in biosignals, Signal pattern classification
Abstract: Telehealthcare has become increasingly popular in clinical practice as a means of providing ubiquitous healthcare through long-term informative interactions and health monitoring. We have delivered a synchronized telehealthcare program since 2009. We have implemented a web-based clinical decision support system with a knowledge-based electrocardiogram (ECG) recognition mechanism as an augmentation service to assist medical practitioners doing decision making in clinical practice. To evaluate the capability and usage limits of this automatic ECG interpretation, the aim of this study was to validate the stability and robustness of proposed mechanism using stress testing through six simulation scenarios. According to experimental results, both of the processing items and processing time augmented steadily by the resource of hardware. Besides, under the cross-validation using 327,058 ECG signals from our telehealthcare program, the recognition classifiers yielded 86.8% accuracy in sinus detection and 88.4% accuracy in atrial fibrillation detection. In the future, this prominent mechanism of automatic ECG interpretation could widely offer high accessibility in the field of medical service. The findings of the present study also encourage and augment further support to implementation of screening and monitoring as part of telehealthcare.

Keywords: Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification, Physiological systems modeling - Signal processing in simulation
Abstract: As a plethora of wearable devices are being introduced, significant concerns exist on the privacy and security of personal data stored on these devices. Expanding on recent works of using electrocardiogram (ECG) as a modality for biometric authentication, in this work, we investigate the possibility of using personal ECG signals as the individually unique source for physical unclonable function (PUF), which eventually can be used as the key for encryption and decryption engines. New signal processing and machine learning algorithms are developed that can learn and extract maximally different ECG features for different individuals and minimally different ECG features for the same individual over time. Experimental results show that the distribution of the intra-subject Hamming distance of extracted ECG features for the same person over time and the distribution of the inter-subject Hamming distance has minimal negligible overlap.