Keywords: Signal pattern classification, Data mining and processing - Pattern recognition, Physiological systems modeling - Signals and systems
Abstract: In this paper, a face-machine interface (FMI) technology is introduced as a new way of communication for patients with amyotrophic lateral sclerosis (ALS). We used surface electromyogram (EMG) and electrooculogram (EOG) to decode the intention of patients with ALS. Our preliminary experimental results demonstrated that the biomedical signals recorded from the human face can be successfully used for communication of the paralyzed.

Keywords: Signal pattern classification, Time-frequency and time-scale analysis - Empirical mode decomposition in biosignal analysis
Abstract: Although human emotional experience plays an important role in our daily lives, its quantitative measurement remains to be one of the most challenging problems in the field of affective sciences. In the present study, we proposed an EEG-based inter-subject neural correlation (ISNC) method to explore human emotional experience during complex and continuous audiovisual video presentation. Our results showed that the correlations between the ISNCs and the subjective emotional ratings were substantially higher than those between the classical power spectra and the ratings, for both the conventional arousal and valence dimensions, as well as the ten discrete positive emotion dimensions. Correspondingly, higher classification accuracies were obtained using ISNC in most cases. Our results suggest that the inter-subject neural correlation approach is a promising candidate for investigating human emotion experiences.

Keywords: Signal pattern classification, Time-frequency and time-scale analysis - Nonstationary processing, Nonlinear dynamic analysis - Biomedical signals
Abstract: In the Minisymposium titled “Recent Progress in Biosignal-Based Human-Computer Interface”, I will introduce technology progress in electromyography (EMG)-based prosthetic hand control and clinical challenges, such as arm position change and donning/doffing, which are important factors associated with performance decrease during use of a prosthetic hand. Also, recent studies addressing the clinical challenges in myoelectric control will be presented and open discussion for developing clinically reliable prosthetic hands will be made.

Keywords: Signal pattern classification, Physiological systems modeling - Multivariate signal processing
Abstract: Brain-computer interface research evolves towards applications in real-life environments. Developing both novel acquisition technology and novel machine learning algorithms mark a crucial step in this process. In this talk, we discuss recent progress in multimodal wireless measurement technology and new machine learning approaches tackling challenges such as feature shifts and severe data contamination.

Keywords: Ablation, Image-guided devices - RF and microwave ablation
Abstract: Cancer remains as one of the leading causes to human death, though medical technology has been much advanced lately. An effective treatment protocol should include the following attributes: (1) directly damaging the cancer cells and their vascular network; (2) destroying the tumor immune-suppression environment and restoring normal immune system surveillance within the body; and (3) stimulating immune response against tumor. In light of this, a new strategy for cancer therapy should synergize the external intervention with a patient’s immune response. Over the past few years, a novel thermo-immune therapy was developed in our Bio-heat and Mass Transfer Laboratory. In this presentation, both the treatment system and clinical studies will be reported, including the protocol, surgical planning and the treatment outcome.

Keywords: Image-guided devices - HIFU, Therapeutic ultrasound, Ablation
Abstract: Efficient energy delivery is increasingly required for clinical HIFU ablation. One typical such approach is volumetric ablation, in which the target volume is divided into different concentric circles (also referred to treatment cells). The focal spots spread uniformly on the circles and each focus is sonicated with the same short time interval. The single-focus and dual-focus patterns were considered to ablate treatment cells with different sizes in simulations. Results show that dual-focus pattern is more suitable for small treatment cells (e.g., 4 mm), while the single-focus pattern is superior for larger cells. Compared with the single-focus pattern, the combination of both patterns can decrease sonication time for volumetric ablation.

Keywords: Therapeutic ultrasound, Image-guided devices - HIFU
Abstract: Tissue undergoes changes in optical properties after ablation but “seeing” those changes at depth in tissue is challenging as light becomes diffuse after propagating a few millimetres resulting in insufficient resolution to see thermal lesions. In acousto-optic (AO) sensing a focused ultrasound beam is used to modulate photons that pass through a small region and by detecting the modulated photons the optical properties from a localised area can be determined. During high-intensity focused ultrasound (HIFU) the ultrasound can be used to modulate the diffuse light during treatment and so the interaction volume coincides with the treatment volume and light modulation can be measured in real-time. A full numerical model captures these effects and can be used to determine the optimal sensing geometry and most appropriate tissues.

Keywords: Ultrasound imaging - Interventional
Abstract: Vulnerable plaques are pointed out as a major cause of cardiovascular disease. However, they are difficult to detect it with conventional intravascular imaging technique due to the devices’ unique limitations. In this article, we develop an intravascular photoacoustic/ultrasound (IVPA-US) catheter and system for arterial structure imaging with lipid identification in plaque based on chemical specificity. We successfully obtained US/PA images of an intravascular metal stent using an IVPA-US catheter.

Keywords: Optical imaging - Coherence tomography, Optical imaging and microscopy - Optical vascular imaging
Abstract: Intravascular optical coherence tomography (OCT) is a catheter-based invasive imaging system. To determine whether coronary revascularization is appropriate and to perform stent evaluation in a catheterization laboratory, OCT examinations are carried out. In this talk, we present the on-going clinical studies to find benefits of OCT-guided stenting using a contour plot and briefly report computational OCT-derived fractional flow reserve in patients. Finally, we discuss the recent developments in a high speed, low noise optical rotary junction and an extended depth-of-focus optical probe.

Keywords: Ultrasound imaging - Vascular imaging, Optical imaging and microscopy - Optical vascular imaging, Multimodal imaging
Abstract: We present advances with optoacoustic and fluorescence imaging, with particular emphasis on recent progress achieved in the intravascular adaptation of those two imaging technologies.

Keywords: Physiological monitoring - Novel methods, Wearable low power, wireless sensing methods, Wearable body sensor networks and telemetric systems
Abstract: Rapid miniaturization and cost reduction of computing, along with the availability of wearable and implantable physiological sensors have led to the growth of human Body Area Network (BAN) formed by a network of such sensors and computing devices. One promising application of such a network is wearable health monitoring where the collected data from the sensors would be transmitted and analyzed to assess the health of a person. Typically, the devices in a BAN are connected through wireless (WBAN), which suffers from energy inefficiency due to the high-energy consumption of wireless transmission. Human Body Communication (HBC) uses the relatively low loss human body as the communication medium to connect these devices, promising order(s) of magnitude better energy-efficiency and built-in security compared to WBAN. In this paper, we demonstrate a health monitoring device and system built using Commercial-Off-The-Shelf (COTS) sensors and components, that can collect data from physiological sensors and transmit it through a) intra-body HBC to another device (hub) worn on the body or b) upload health data through HBC-based human-machine interaction to an HBC capable machine. The system design constraints and signal transfer characteristics for the implemented HBC-based wearable health monitoring system are measured and analyzed, showing a very reliable connectivity with >8x power savings compared to Bluetooth low-energy (BTLE).

Keywords: Bio-electric sensors - Sensor systems, Bio-electric sensors - Sensing methods, Chemo/bio-sensing - Chemical sensors and systems
Abstract: A novel biosensing system capable of simultaneously sensing glucose and powering portable electronic devices such as a digital glucometer is described. The biosensing system consists of enzymatic glucose biofuel cell bioelectrodes functionalized with pyrolloquinoline quinone glucose dehydrogenase (PQQ-GDH) and bilirubin oxidase (BOD) at the bioanode and biocathode, respectively. A dual-stage power amplification circuit is integrated with the single biofuel cell to amplify the electrical power generated. In addition, a capacitor circuit was incorporated to serve as the transducer for sensing glucose. The open circuit voltage of the optimized biofuel cell reached 0.55 V, and the maximum power density achieved was 0.23 mW/ cm2 at 0.29 V. The biofuel cell exhibited a sensitivity of 0.312 mW/mM.cm2 with a linear dynamic range of 3 mM – 20 mM glucose. The overall self-powered glucose biosensor is capable of selectively screening against common interfering species, such as ascorbate and urate and exhibited an operational stability of over 53 days, while maintaining 90 % of its activity. These results demonstrate the system’s potential to replace the current glucose monitoring devices that rely on external power supply, such as a battery.

Keywords: Bio-electric sensors - Sensing methods, Physiological monitoring - Modeling and analysis, Physiological monitoring - Instrumentation
Abstract: In an ongoing project for electrical impedance-based needle guidance we have previously showed in an animal model that intraneural needle positions can be detected with bioimpedance measurement. To enhance the power of this method we in this study have investigated whether an early detection of the needle only touching the nerve also is feasible. Measurement of complex impedance during needle to nerve contact was compared with needle positions in surrounding tissues in a volunteer study on 32 subjects. Classification analysis using Support-Vector Machines demonstrated that discrimination is possible, but that the sensitivity and specificity for the nerve touch algorithm not is at the same level of performance as for intra-neuralintraneural detection.

Keywords: Bio-electric sensors - Sensing methods
Abstract: We compared previously carbon/salt/adhesive (CSA) electrodes with Ag/AgCl electrodes for surface electromyography (sEMG) signals collection. We found no differences in amplitude, but CSA electrodes exhibited a significantly better response to noise and motion artifacts. However, the carbon component may not be needed, and the salt/adhesive (SA) mixture might be as good as CSA for such a task. Either CSA or SA mixtures have the potential to provide the unique advantages of having longer (theoretically infinite) shelf life and potentially lower cost, compared to the gold standard Ag/AgCl hydrogel electrodes. In order to determine if carbon contribution is necessary for effective sEMG measuring capabilities the mixture, the functionality of SA electrodes utilizing different levels of salt concentration were compared to the capabilities of CSA electrodes. The levels consisted of 10%, 15%, and 25% salt concentration. Six subjects have been recruited so far to collect simultaneous recordings of sEMG signals using CSA and SA electrodes, side-by-side on triceps brachii, tibial anterior muscles, biceps brachii and quadriceps femoris. For all three levels of salt concentration in the SA electrodes, high correlation was found to the CSA electrodes on the estimated linear envelopes, RMS envelope and power spectrum density. Furthermore, no significant differences in amplitude, compared to CSA electrodes, were found for the three concentrations. Based on signal-to-noise and signal-to-motion measures on the preliminary data set, it seems like adding carbon to the mixture improves the response to motion, but impairs the noise corruption of the sEMG signals.

Keywords: Bio-electric sensors - Sensor systems, Portable miniaturized systems, Physical sensors and sensor systems - Optical and photonic Sensors and systems
Abstract: Measurement of Luteinizing hormone (LH) levels is of great importance in guidance for pregnancy, diagnosis of ovarian diseases and evaluation of clinical effect. Gold immuno- chromatographic strip(GICS) assay is a rapid, simple, single- copy and on-site method. Quantitative interpretation of GICS can provide more information than the traditional qualitative or semi-quantitative strip assay. This paper aims to develop a quantitative determination method for GICS based on smart-phone. First, smart-phone was used to acquire GICS image. Then, we applied the canny edge detection operator to extract reading window of GICS, and the fuzzy c-means (FCM) clustering algorithm to extract the test and control lines in the reading window. In order to reduce environmental interference，luminance-compensation based on color constancy algorithms was applied. Finally, the property of the developed quantitative method is demonstrated by the detection of LH sample and clinical serum sample. Experimental results revealed that this method could achieve a low detection limit of 1.0 mIU/mL in a linear range from 1.0 to 125.0 mIU/mL. Furthermore, the proposed method could be used for the determination of clinical serum samples and its corresponding correlation coefficients were 0.964. Results showed that this novel method could be an effective tool for the rapid quantitative determination of LH.

Keywords: Bio-electric sensors - Sensing methods, Bio-electric sensors - Sensor systems
Abstract: Recently, highly flexible and soft pressure distribution imaging sensor is in great demand for tactile sensing, gait analysis, ubiquitous life-care based on activity recognition, and therapeutics. In this study, we integrate the piezo-capacitive and piezo-electric nanowebs with the conductive fabric sheets for detecting static and dynamic pressure distributions on a large sensing area. Electrical impedance tomography (EIT) and electric source imaging are applied for reconstructing pressure distribution images from measured current-voltage data on the boundary of the hybrid fabric sensor. We evaluated the piezo-capacitive nanoweb sensor, piezo-electric nanoweb sensor, and hybrid fabric sensor. The results show the feasibility of static and dynamic pressure distribution imaging from the boundary measurements of the fabric sensors.

Keywords: Teaching design, Instruction and learning, Novel approaches to BME education
Abstract: The role of Problem Based Learning (PBL) is relative clear in domains such as medicine but its efficacy in bio-medical engineering and engineering in general is as yet less certain. To clarify the role of PBL in engineering, a 3 day workshop was conducted for senior Brazilian engineering academics where they were given the theory and then an immersive PBL experience. One major purpose for running this workshop was for them to identify suitable courses where PBL could be considered. During this workshop, they were split in teams and given a diverse range of problems. At the conclusion of the workshop, a quantifiable survey was conducted and the results show that PBL can deliver superior educational outcomes providing the student group is drawn from the top 5% of the year 12 students, and that significantly higher resources are made available. Thus, any proposed PBL program in engineering must be able to demonstrate that it can meet these requirements before it can move forward to implementation.

Keywords: BME and global health, Biomedical engineering curricula, BME undergraduate research
Abstract: Under the auspices of the MIT D-Lab, a team of graduate students has instructed a course since 2008 that centers around teaching low-cost prosthetic design for resource-constrained environments. Recently, the course has evolved into a fully immersive design experience that pairs student teams with real-life international stakeholders and industry partners. Following this structure, projects initiated as part of the course have been tested at field sites around the world, stimulated further research, advanced student careers, raised additional grant money, and generated peer-reviewed publications and intellectual property.

Keywords: Novel approaches to BME education, Teaching design
Abstract: Biomedical Engineering (BME) bachelor education aims to train qualified engineers who devote themselves to addressing biological and medical problems by integrating the technological, medical and biological knowledge. Design thinking and teamwork with other disciplines are necessary for biomedical engineers. In the current biomedical engineering education system of Shanghai University (SHU), however, such design thinking and teamwork through a practical project is lacking. This paper describes a creative “joint assignment” project in Shanghai University, China, which has provided BME bachelor students a two-year practical experience to work with students from multidisciplinary departments including sociology, mechanics, computer sciences, business and art, etc. To test the feasibility of this project, a twenty-month pilot project has been carried out from May 2015 to December 2016. The results showed that this pilot project obviously enhanced competitive power of BME students in Shanghai University, both in the capabilities of design thinking and teamwork.

Keywords: Novel approaches to BME education, Teaching design
Abstract: The “joint assignment” is a creative bachelor education project for Biomedical Engineering (BME) in Shanghai University (SHU), China. The objective of this project is to improve students’ capabilities in design thinking and teamwork through practices in the process of the design and development of complex medical product. As the first step, a pilot project “design and development of intelligent nursing bed” was set up in May 2015. This paper describes details of how project organization and management, various teaching methods and scientific evaluation approaches were achieved in this pilot project. For example, a method containing one main line and four branches is taken to manage the project and “prototyping model” was used as the main research approach. As a result a multi-win situation was achieved. The results showed, firstly 62 bachelor students, including 16 BME students were well trained. They improved themselves in use of practical tools, communication skills and scientific writing; Secondly commercial companies received a nice product design on intelligent nursing bed, and have been working on industrializing it; Thirdly the university and associated schools obtained an excellent practical education experience to supplement traditional class education; Fourthly and most importantly, requirements from end-users will be met. The results also showed that the “joint assignment” task could be one significant component in BME bachelor education.

Keywords: Novel approaches to BME education, Instruction and learning, Teaching design
Abstract: Abstract--- Epidural and spinal anesthesia are mostly performed “blind” without any medical imaging. Currently, training of these procedures is performed on human specimens, virtual reality systems, manikins and mostly in clinical practice supervised by a professional. In this study a novel hybrid, low-cost patient simulator for the training of needle insertion into the epidural space was designed. The patient phantom provides a realistic force feedback comparable with biological tissue and enables sensing of the needle tip position during insertion. A display delivers the trainee a real-time feedback of the needle tip position.

Keywords: Novel approaches to BME education, Teaching design, Instruction and learning
Abstract: During vertebral surgery, misplaced pedicle screws can harm vital neural and vascular structures. Haptic distinction between cortical and cancellous bone structures is therefore essential for correct screw placement. This tactile experience during pedicle screw placement can be obtained by training on human or animal specimens albeit expensive or ethically questionable. In this study, novel synthetic vertebrae were evaluated within a hybrid simulator to provide realistic haptics for the training of spine surgeries. Synthetic vertebrae made of calcium powder-based composites and imitating both, cancellous and cortical bone, were custom made. Mechanical properties of synthetic surrogates were validated for pedicle screw placement and cement augmentation and were compared with those obtained on human vertebrae and insertion torques were analyzed. In human vertebrae pedicle screw torque measurements resulted in mean torque slopes of 82pm33Nm/m. Calcium carbonate-based materials achieved lower torques than the human bone whereas calcium phosphate-based bone surrogates showed comparable results. A further differentiation of the calcium phosphate-based vertebrae revealed, that synthetic vertebrae with lower amounts of blowing agent, achieved suitable torques (83 pm 28Nm/m) in comparison to the human reference (p = 0.387). Cement application and subsequent fluoroscopy images confirmed, that the cancellous core of the synthetic vertebrae enabled cement augmentation. In conclusion, our findings suggest, that the artificial bone samples mimic the properties of human bone during pedicle screw placement and cement augmentation and are therefore suitable as synthetic vertebrae in a hybrid surgical simulator.

Keywords: Neuromuscular systems - EMG processing and applications, Neuromuscular systems - Computational modeling, Neuromuscular systems - Learning and adaption
Abstract: We study in this work the feasibility of early prediction of hand movement based on sEMG signals to overcome the time delay issue of the conventional classification. Opposed to the classification task, the objective of early prediction is to predict a hand movement that is going to occur in the future given the information up to the current time point. The ability of early prediction may allow a hand prosthesis control system to compensate for the time delay and, as a result, improve the usability. Experimental results on the Ninapro database show that we can predict up to 300 ms ahead in the future while the prediction accuracy remains very close to that of the standard classification, i.e. it is just marginally lower. Furthermore, historical data prior the current time window is shown to be very important to improve performance, not only for the prediction but also the classification task.

Keywords: Neuromuscular systems - EMG processing and applications, Neuromuscular systems - Locomotion, Human performance - Gait
Abstract: During cyclic movements, the number of muscle activations and their timing are different from cycle to cycle. In a previous study, the CIMAP algorithm was proposed for grouping cycles showing similar EMG activation intervals, using dendrogram clustering. Even if the algorithm demonstrated good performances on a healthy population, the intra-cluster variability decreased when applied to datasets from pathological subjects. In this work we propose an optimized version of the CIMAP, comparing the performances of 8 different combinations of parameters used for the dendrogram construction. The cut-off point is also modified. The new and the original version of the algorithm are compared, in terms of intra-cluster variability, considering a population of 60 subjects, both healthy and pathological. The results show that the new CIMAP allows for obtaining clusters with lower variability with respect to the original version of the algorithm (p < 0.001).

Keywords: Neuromuscular systems - EMG processing and applications
Abstract: In recent years, low-cost, low-power myoelectric control systems such as the Myo armband from Thalmic Labs have become available and unlocked tremendous possibilities for myoelectric controlled applications. However, due to the embedded system constraints, such sEMG control devices typically samples sEMG signals at a lower frequency. It is in doubt whether existing sEMG feature extraction methods are still valid on such low-resolution sEMG data. In addition, the feature extraction algorithms implemented on embedded devices must have low computational complexity in order to meet the real-time requirement. This paper aims to investigate effective features for low-resolution EMG pattern recognition. In particular, a set of novel computational efficient space-domain (SD) features (referred to as simple SD (SSD) features) have been developed to exploit the spatial relationships of sEMG signals recorded from the sensor array on the Myo armband. The proposed SSD feature set was evaluated with a linear discriminant analysis (LDA)-based classifier on a 9-gesture dataset. The experimental results indicate that using the SSD features increased the classification accuracy by 5% compared to using Hudgins’ time-domain features.

Keywords: Neuromuscular systems - Learning and adaption, Human performance - Gait, Neuromuscular systems - Postural and balance
Abstract: Controlling center of mass (COM) position and velocity within a dynamic base of support is essential for gait stability. This skill is often compromised following neurologic injury creating a need to develop effective interventions to enhance gait stability. A movement augmentation paradigm applied to walking could potentially be used to improve control of COM dynamics. We have developed a cable robot system, the Agility Trainer, to apply continuous frontal-plane forces to the pelvis during treadmill walking. This cable robot system uses a set of series elastic actuators powered by linear motors to create bilateral forces. Here we use the Agility Trainer to create a negative viscosity force field proportional to the subject’s lateral velocity. Two healthy young subjects performed two 10-minute walking trials, Baseline and Negative Viscosity. During the first minute of walking in the Negative Viscosity field participants’ lateral COM motion became irregular when compared to the rhythmic sinusoidal motion observed during Baseline walking. By the 10th minute of walking in the Negative Viscosity field the participants had adapted their gait patterns, decreasing their variation in peak lateral COM speed each stride. These results demonstrate that it is feasible to use the Agility Trainer to safely apply a movement augmentation paradigm to human walking.

Keywords: Neuromuscular systems - Postural and balance, Mechanics of locomotion and balance, Joint biomechanics
Abstract: Dynamic ankle stiffness, which defines the relationship between joint position and the torque acting about it, determines the joint resistance to external perturbations. The contribution of joint stiffness to stance control is usually assumed to be constant; however, the stiffness may be modulated due to the ongoing sway during stance. This paper describes the results of the direct estimation of ankle intrinsic stiffness and its modulation with background torque during stance. Three subjects stood on a standing apparatus, while subjected to pulse perturbations of ankle position. The angle of center of mass, and ankle torque and angle were measured and individual torque responses to each position perturbation were analyzed. Each response lasted 90 milliseconds, where the mean of the torque record in the first 25 milliseconds was considered as its background torque. Ensemble of the responses with similar background torques were used to estimate the joint intrinsic stiffness. The estimated parameters showed that elastic element of the ankle intrinsic stiffness varied significantly (by a factor of 3) with background torque, i.e. the higher the background torque, the higher the stiffness. Thus ankle stiffness is not constant but varies greatly with postural sway.

Keywords: Neural interfaces - Implantable systems, Neural stimulation, Sensory neuroprostheses
Abstract: Electrode-based neural prosthetic implants that target the neocortex have the potential to treat a wide range of neurological and psychiatric disorders. However, their effectiveness has been limited due to their inability to selectively activate specific sub-populations of cortical neurons as well as the complex biological reactions that diminish the viability of electrodes over time. Magnetic stimulation is thought to overcome many of these limitations but coils small enough to be cortically implanted were not thought capable of neuronal activation. Here, we describe a new micro-coil designed for cortical implantation and demonstrate its effectiveness via a series of in vitro and in vivo experiments. A 50 x 100 µm sized micro-coil was designed and fabricated using silicon micro-fabrication technologies. In vitro experiments revealed that cortical pyramidal neurons in mouse brain slices could be strongly activated by the micro-coil stimulation without simultaneously activating passing axons. In vivo animal experiments showed that the micro-coils could be safely implanted into mouse cortex and could drive neural circuits in living animals. These results suggest that micro-coil implants may be an effective and reliable alternative to the existing electrode-based cortical implants.

Keywords: Sensory neuroprostheses - Visual, Neural interfaces - Microelectrode technology, Brain-computer/machine interface
Abstract: We have developed a bionic vision system based on wirelessly powered implants for the primary visual cortex, each with 43 microelectrodes. There have been many engineering and medical challenges to overcome to produce a long lasting, biocompatible, safe and effective implant. These issues will be explored in this presentation. The design and timing of the first-in-human trial will be discussed.

Keywords: Sensory neuroprostheses - Somatosensory, Brain-computer/machine interface, Neural stimulation
Abstract: Intracortical microstimulation of the somatosensory cortex offers the potential for creating a sensory neuroprosthesis to restore tactile sensations. We recently demonstrated that microstimulation within the hand area of somatosensory cortex of a person with long-term spinal cord injury evokes tactile sensations perceived as originating from various locations on his own hand. A key issue for developing a reliable neuroprosthesis using this approach is percept stability. Here we measure the number of electrodes that elicit responses during a standardized survey of electrode function over 20 months. We found that the number of electrodes that were detected and described increased significantly, suggesting that this approach might be viable for long-term sensory restoration.

Keywords: Systems modeling - Clinical applications of biological networks, Systems modeling - Decision making
Abstract: Heart transplantation has enabled to extend the median survival rate to 12 years for patients with end-stage heart diseases. This operation is unfortunately limited by the availability of donor organs and patients have to wait on average about 200 days in a waiting list before being operated. This waiting time varies considerably across the patients. In this paper, we studied the outcome for patients entering a transplantation waiting list using deep learning techniques. We implemented a model in the form of two-layer neural networks and we predicted the outcome as: still waiting, transplanted, or dead in the waiting list at three different time points: 180 days, 365 days, and 730 days. As data source, we used the United Network for Organ Sharing (UNOS) registry, where we extracted adult patients (> 17 years) from January 2000 to December 2011. We trained our model using the Keras framework and we report F1 macro scores of respectively 0.674, 0.680, and 0.680 compared to a baseline of 0.271. We also applied a backward elimination procedure, using our neural network, to extract the 10 most significant parameters predicting the patient status for the three different time points.

Keywords: Data-driven modeling
Abstract: Recent progress in biosensor technology and wearable devices has created a formidable opportunity for remote healthcare monitoring systems as well as real-time diagnosis and disease prevention. The use of data mining techniques is indispensable for analysis of the large pool of data generated by the wearable devices. Deep learning is among the promising methods for analyzing such data for healthcare applications and disease diagnosis. However, the conventional deep neural networks are computationally intensive and it is impractical to use them in real-time diagnosis with low-powered on-body devices. We propose Staged Inference using Conditional Deep Learning (SICDL), as an energy efficient approach for creating healthcare monitoring systems. For smart diagnostics, we observe that all diagnoses are not equally challenging. The proposed approach thus decomposes the diagnoses into preliminary analysis (such as healthy vs unhealthy) and detailed analysis (such as identifying the specific type of cardio disease). The preliminary diagnosis is conducted real-time with a low complexity neural network realized on the resource-constrained on-body device. The detailed diagnosis requires a larger network that is implemented remotely in cloud and is conditionally activated only for detailed diagnosis (unhealthy individuals). We evaluated the proposed approach using available physiological sensor data from Physionet databases, and achieved 38% energy reduction in comparison to the conventional deep learning approach.

Keywords: Systems modeling - Patient stratification, Systems modeling - Decision making
Abstract: Metabolomic data analysis presents a unique opportunity to advance our understanding of osteosarcoma, a common bone malignancy for which genomic and proteomic studies have enjoyed limited success. One of the major goals of metabolomic studies is to classify osteosarcoma in early stages, which is required for metastasectomy treatment. In this paper we subject our metabolomic data on osteosarcoma patients collected by the SJTU team to three classification methods: logistic regression, support vector machine (SVM) and random forest (RF). The performances are evaluated and compared using receiver operating characteristic curves. All three classifiers are successful in distinguishing between healthy control and tumor cases, with random forest outperforming the other two for cross-validation in training set (accuracy rate for logistic regression, support vector machine and random forest are 88%, 90% and 97% respectively). Random forest achieved overall accuracy rate of 95% with 0.99 AUC on testing set.

Keywords: Data-driven modeling, Modeling of cell, tissue, and regenerative medicine - Cell movement, cell migration, cell motility, Modeling of cell, tissue, and regenerative medicine - Tissue formation
Abstract: Ductal carcinoma in situ (DCIS) is the most commonly diagnosed form of non-invasive breast cancer, constituting 20% of all new breast cancer cases in the United States. Although non-invasive, DCIS is usually treated surgically through resection. Interestingly, long-term survival studies have shown that patient survival rates are not significantly impacted by the type or resection, indicating that increased breast conservation through minimized surgical resection may indeed be possible. This requires a greater understanding of disease development, so that clinicians may more accurately determine surgical margins which minimize patient impact while maintaining survival rates. To this end, we have developed a three-dimensional, lattice-free multiscale agent based model of DCIS designed to help quantify ductal invasion rates, in order to allow clinicians to better estimate disease age and extent of invasion, and to predict surgical margins based on parameters obtainable from non-invasive testing (i.e., mammography). Here, we present the model development to date, and discuss some preliminary results.

Keywords: Translational biomedical informatics - Knowledge modeling, Computational modeling - Biological networks, Systems modeling - Clinical applications of biological networks
Abstract: The purple or blue coloration of hands and feet, known as peripheral cyanosis, can represent one of the initial signs of potentially life threatening medical conditions. Consequently, procedures aimed at its early detection and interpretation can help health-care professionals to select the appropriate treatment for these conditions. The effectiveness of such procedures, in turn, depends on the correct assessment of the biophysical processes responsible for eliciting this abnormal skin appearance. However, despite the diverse body of existing clinical research involving cyanosis, the interplay between physiological changes and the optical phenomena leading to cyanotic responses remains not fully understood. In this paper, we methodically examine this interplay through controlled in silico experiments. Among other relevant aspects, the results of our experiments demonstrate that Rayleigh scattering, a light attenuation phenomenon overlooked by previous studies on peripheral cyanosis, plays a pivotal role in the manifestation of cyanotic chromatic attributes. We believe that the insights derived from our experiments can contribute to the development of more effective protocols for the screening of medical conditions associated with peripheral cyanosis etiology.

Keywords: Organs and medical devices - Multiscale modeling and the physiome, Models of organs and medical devices - Inverse problems in biology, Computational modeling - Analysis of high-throughput systems biology data
Abstract: SMARTool aims to the development of a clinical decision support system (CDSS) for the management and stratification of patients with coronary artery disease (CAD). This will be achieved by performing computational modeling of the main processes of atherosclerotic plaque growth. More specifically, computed tomography coronary angiography (CTCA) is acquired and 3-dimensional (3D) reconstruction is performed for the arterial trees. Then, blood flow and plaque growth modeling is employed simulating the major processes of atherosclerosis, such as the estimation of endothelial shear stress (ESS), the lipids transportation, low density lipoprotein (LDL) oxidation, macrophages migration and plaque development. The plaque growth model integrates information from genetic and biological data of the patients. The SMARTool system enables also the calculation of the virtual functional assessment index (vFAI), an index equivalent to the invasively measured fractional flow reserve (FFR), to provide decision support for patients with stenosed arteries. Finally, it integrates modeling of stent deployment. In this work preliminary results are presented. More specifically, the reconstruction methodology has mean value of Dice Coefficient and Hausdorff Distance is 0.749 and 1.746, respectively, while low ESS and high LDL concentration can predict plaque progression.

Keywords: Coronary artery disease, Cardiovascular regulation - Heart rate variability, Coronary blood flow
Abstract: This paper presents a technique for coronary artery disease (CAD) detection through photoplethysmography (PPG). This work is aimed at developing a non-invasive, inexpensive screening technique suitable for home monitoring. Time domain analysis of PPG signal and its second derivative has been carried out to extract distinguishing features. Support Vector Machine based classifier has been used to classify CAD patients. ICU patient data from MIMIC-II dataset has been used for performance evaluation. Sensitivity of 85% and specificity of 78% has been achieved for the analysed data.

Keywords: Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: Atrial fibrillation (AF) is one of the major causes of stroke, heart failure, sudden death, and cardiovascular morbidity and the most common type of arrhythmia. Its diagnosis and the initiation of treatment, however, currently requires electrocardiogram (ECG)-based heart rhythm monitoring. The photoplethysmogram (PPG) offers an alternative method, which is convenient in terms of its recording and allows for self-monitoring, thus relieving clinical staff and enabling early AF diagnosis. We introduce a PPG-based AF detection algorithm using smartphones that has a low computational cost and low memory requirements. In particular, we propose a modified PPG signal acquisition, explore new statistical discriminating features and propose simple classification equations by using sequential forward selection (SFS) and support vector machines (SVM). The algorithm is applied to clinical data and evaluated in terms of receiver operating characteristic (ROC) curve and statistical measures. The combination of Shannon entropy and the median of the peak rise height achieves perfect detection of AF on the recorded data, highlighting the potential of PPG for reliable AF detection.

Keywords: Cardiovascular and respiratory signal processing - Cardiovascular signal processing, Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability, Cardiovascular regulation - Heart rate variability
Abstract: Photoplethysmography(PPG) as a non-invasive tool for monitoring various cardiovascular parameters, has become popular due to the ease of wearable integration and pervasive nature. Due to unobtrusive nature of sensor placement at wrist, smartwatches and wrist based fitness bands have gained popularity. However, any movement of the wrist along with frequent loose contacts significantly corrupts the PPG signal. Reliable peak detection from the corrupted PPG signal is essential for any further processing, as many physiological quantities such as heart rate variability (HRV) depends on the peak-to-peak distances in the PPG signal, known as the RR Series. This paper attempts to provide a robust algorithm for peak detection in noise & motion artefact corrupted PPG signals. The algorithm consists of steps to remove the baseline drift in the PPG signal using wavelet filtering and trend removal and subsequent peak detection using autcorrelation for each pseudo-periodic segment of the signal. The validation of the method is done by comparing the PPG peaks detected by the algorithm with RR series extracted from simultaneously captured ECG signal.

Keywords: Cardiovascular and respiratory signal processing - Heart Rate and Blood Pressure Variability, Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: Cardiovascular Diseases (CVDs) cause a very large number of casualties around the world every year and cardiac arrhythmias contribute to significant proportion of CVD related deaths. Bedside cardiac activity monitors in hospitals are based on electrocardiogram (ECG) processing and are known to produce too many false alarms. Moving beyond bedside care, ECG is not very suitable for use in wearable devices. Photoplethysmography (PPG) on the other hand provides an inexpensive and more wearable device-friendly alternative. This work presents a technique to detect life threatening arrhythmias using only PPG waveforms. PhysioNet Challenge 2015 data is used to detect five types of arrhythmias namely, tachycardia, bradycardia, asystole, ventricular tachycardia and ventricular fibrillation. Novel technique is employed to assign pulse quality index to every PPG pulse and highest quality portion of the signal is used for detection. Results indicate that PPG provides a viable alternative for conventional ECG based detection. An overall true positive rate (TPR) of 93% was achieved with true negative rate (TNR) of 53.78% suggesting that PPG is a viable option for arrhythmia detection.

Keywords: Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: Photoplethysmography (PPG) is one of the key technologies for unobtrusive physiological monitoring, with ongoing attempts to use it in several medical fields, ranging from night to night sleep analysis to continuous cardiac arrhythmia monitoring. However, the PPG signals are susceptible to be corrupted by noise and artifacts, caused, e.g., by limb or sensor movement. These artifacts affect the morphology of PPG waves and prevent the accurate detection and localization of beats and subsequent cardiovascular feature extraction. In this paper a new algorithm for beat detection and pulse quality assessment is described. The algorithm segments the PPG signal in pulses, localizes each beat and grades each segment with a quality index. The obtained index results from a comparison between each pulse and a template derived from the surrounding pulses, by mean of dynamic time warping barycenter averaging. The quality index is used to discard corrupted pulse beats. The algorithm is evaluated by comparing the detected beats with annotated PPG signals and the results are published over the same data. The described method achieves an improved sensitivity and a higher predictive value.

Keywords: Sleep - Obstructive sleep apnea, Cardiovascular and respiratory signal processing - Complexity in cardiovascular or respiratory signals
Abstract: This paper presents a study on identifying sleep apnea using the photoplethysmography (PPG) measurements, which is obtained from a low-cost, non-invasive and a convenient-to-use sensor. Using a database of polysomnogram records of 52 patients, the heart rate and breathing effort information is derived from the PPG measurements and then features are extracted and applied to a classification model to detect sleep apnea. The apnea classification is carried out for each one-minute segment of the signal. Pulse oximetry (SpO2) measurements from the same database are also used in the classification process for comparison and in combination with the PPG results. The results show that both the heart rate and respiratory effort information derived from the PPG signal were able to detect apneic segments with some success. The best apnea classification performance was obtained when we combined the SpO2 features and the PPG features.

Keywords: Wearable sensor systems - User centered design and applications, Physiological monitoring - Novel methods
Abstract: As the world population is growing toward an aging society, elderly fall becomes a serious problem. Automatic fall detection and alert systems could shorten their waiting time after a fall and mitigate its physical and mental negative consequences. This work proposes a method that integrates a 3-axis accelerometer and a barometer on a wrist-worn device for the fall detection task. The method focuses on the use of noisy signals from a barometer in both pre-processing steps and feature extractions. A use of free falling events to address the lack of training data in a learning process is also explored. An evaluation using simulated falls and various activities shows a high classification performance except for a few false alarms occurring when sitting on the floor from a standing pose.

Keywords: Physiological monitoring - Modeling and analysis, Wearable wireless sensors, motes and systems
Abstract: The objective of this study was to investigate the validity of an objective gait measure for assessment of different motor states of advanced Parkinson’s disease (PD) patients. Seven PD patients performed a gait task up to 15 times while wearing sensors on their upper and lower limbs. Each task was performed at specific points during a test day, following a single dose of levodopa-carbidopa. At the time of the tasks the patients were video recorded and three movement disorder experts rated their motor function on three clinical scales: a treatment response scale (TRS) that ranged from -3 (very bradykinetic) to 0 (on) to +3 (very dyskinetic), a dyskinesia score that ranged from 0 (no dyskinesia) to 4 (extreme dyskinesia), and a bradykinesia score that ranged from 0 (no bradykinesia) to 4 (extreme bradykinesia). Raw accelerometer and gyroscope data of the sensors were processed and analyzed with time series analysis methods to extract features. The utilized features quantified separate limb movements as well as movement symmetries between the limbs. The features were processed with principal component analysis and the components were used as predictors for separate support vector machine (SVM) models for each of the three scales. The performance of each model was evaluated in a leave-one-patient out setting where the observations of a single patient were used as the testing set and the observations of the other 6 patients as the training set. Root mean square error (RMSE) and correlation coefficients for the predictions showed a good ability of the models to map the sensor data into the rating scales. There were strong correlations between the SVM models and the mean ratings of TRS (0.79; RMSE=0.70), bradykinesia score (0.79; RMSE=0.47), and bradykinesia score (0.78; RMSE=0.46). The results presented in this paper indicate that the use of wearable sensors when performing gait tasks can generate measurements that have a good correlation to subjective expert assessments

Keywords: Wearable low power, wireless sensing methods, Physiological monitoring - Instrumentation, Physical sensors and sensor systems - New sensing techniques
Abstract: Manual pulse palpation is the common procedure to assess pulse in unconscious patients. This is an error prone procedure during cardiopulmonary resuscitation and therefore automatic pulse detection techniques are being investigated. Accelerometry is an interesting sensing modality for this type of applications. However, accelerometers are highly prone to movement artifacts. Hence, one challenge in designing a solution using accelerometers is to handle motion artifacts properly. In this paper we investigate computationally simple features and classifier to capture movement artifacts in accelerometer signals acquired from the carotid. In particular, based on data obtained from health subjects we show that it is possible to use simple features to achieve an artifact detection sensitivity and specificity higher than 90%.

Keywords: Portable miniaturized systems, Wearable body sensor networks and telemetric systems, Wearable sensor systems - User centered design and applications
Abstract: The pressure field that exists between the foot and the supporting surface is identified as the foot plantar pressure. The information obtained from foot plantar pressure measurements has useful applications that include diagnosis of gait disturbances, optimization of footwear design, sport biomechanics and prevention of injury. Using wearable technology to measure foot plantar pressure continuously allows the collection of comprehensive real-life data sets while interfering minimally with the subject’s daily activities. This paper presents the design of a wearable device to measure foot plantar pressure. Mechanical and electrical design considerations as well as data analysis are discussed. A pilot study involving 20 physically fit volunteers (15 males and 5 females, ageing from 20 – 45) performing a variety of physical activities (such as standing, walking, jumping and climbing up and down stairs) illustrate the potential of the device in terms of its wearability, and suitability for unobtrusive long-term monitoring.

Keywords: Physiological monitoring - Modeling and analysis, Wearable wireless sensors, motes and systems
Abstract: Continuous, automated monitoring of Parkinson’s Disease (PD) symptoms would provide clinicians with more information to understand their patients' disease progression and adjust treatment protocols, thereby improving care. Collecting precisely labeled data for Parkinson's symptoms, such as tremor, is difficult. Therefore, algorithms for monitoring should only require weakly-labeled training data. In this paper, we evaluate five standard weakly-supervised algorithms and propose a ``stratified'' version of three of the algorithms, which take advantage of knowing the approximate amount of tremor within each segment. We analyze PD tremor detection performance as training segments increase in length from 30 seconds to 10 minutes, and labels thereby become less precise. As segment length increases to 10 minutes, standard algorithms are not able to discriminate tremor from non-tremor. However, our stratified algorithms, which make use of more nuanced labels, show little decrease in performance as segment length increases.

Keywords: Wearable sensor systems - User centered design and applications, Wearable low power, wireless sensing methods, Integrated sensor systems
Abstract: In this paper, we present the evaluation of a new smart shoe capable of performing gait analysis in real time. The system is exclusively based on accelerometers which minimizes the power consumption. The estimated parameters are activity class (rest/walk/run), step cadence, ground contact time, foot impact (zone, strength, and balance), forward distance, and speed. The different parameters have been validated with a customized database of 26 subjects on a treadmill and video data labeled manually. Key measures for running analysis such as the cadence is retrieved with a maximum error of 2%, and the ground contact time with an average error of 3.25%. The classification of the foot impact zone achieves a precision between 72% and 91% depending of the running style. The presented algorithm has been licensed to ICON Health & Fitness Inc. for their line of wearables under the brand iFit.

Keywords: Drug delivery systems and carriers - Peptide and protein drug delivery, Micro and Nano formulation - Nanotechnology/Nanoparticles, Drug formulaion - Suspensions
Abstract: A liquid-gallium based nanoformulation is developed for co-delivery of a membrane-associated cytokine (TRAIL) and an intracellular-acting small-molecule drug (CDDP) to overcome drug-resistance in cancer cells. This fully biodegradable nanocarrier could intracellularly transform into therapeutic agent Ga (III) ions to facilitate to reverse drug resistance and to promote anticancer efficacy.

Keywords: Micro and Nano formulation - Nanotechnology/Nanoparticles, Drug delivery systems and carriers - Liposomes/Lipid-based drug delivery, Pharmaceutical engineering
Abstract: Abstract—The rapid growth of cancer cells as well as the distorted blood vessels cause low oxygen content in solid tumors. Tumor hypoxia can bring about the chemotherapeutics resistance. To overcome hypoxia-induced chemoresistance, we introduced a cancer cell membrane-modified smart nanosystem with a polymeric core encapsulated natural hemoglobin (Hb) and anti-cancer drug doxorubicin (DOX) for efficient chemotherapy.

The well-defined DOX/Hb-loaded and cancer cell membrane-coated nanoparticles (DHCNPs) were constructed from two components, the first is a DOX/Hb-loaded polymeric NPs prepared by single-step assembly. The second is a cancer cell membrane shell extracted from MCF-7 cancer cells and further coated onto pre-formed polymeric NPs by extrusion. The DHCNPs availably retained the cancer cell adhesion molecules (e.g., EpCAM, N-cadherin, and galectin-3) and reserved the oxygen-carrying capacity of Hb [1,2]. Benefited from the homologous targeting of adhesion molecules [3], DHCNPs significantly promoted the DOX endocytosis of MCF-7 breast cancer cells, and enhanced homologous-targeting tumor accumulation. Meanwhile, the oxygen intervention of DHCNPS noticeably downregulated the HIF-1α expression and eventually reduced exocytosis of DOX by the decreased expression of p-glycoprotein. The in vitro MTS assay showed that hypoxic MCF-7 cancer cells treated with DHCNPs caused about 80% cell death, which was 3.2 fold than that of treated with free DOX. DHCNPs treated mice achieved 100.0% survival rate in the in vivo therapeutic experiment. By the interactions of homologous target recognition and oxygen adjustment, the DHCNPs showed highly effective chemotherapy in vitro and in vivo. The biomimetic DHCNPs exhibited the attractive prospects in drug-resistant cancer therapy.

Keywords: Micro and Nano formulation - Nanotechnology/Nanoparticles, Drug formulaion - Polymers, Drug release and solubility - Controlled/Sustained/Modified release
Abstract: Unimolecular micelles, formed by single/individual multi-arm star amphiphilic block copolymers, possess outstanding in vivo stability due to their covalent nature. Thus, they are a desirable drug nanocarrier for targeted cancer therapy. We have previously demonstrated that unimolecular micelles conjugated with tumor-targeting ligands can greatly enhance the tumor accumulation and anticancer efficacy of nanoparticles in regular xenograft mouse models. Recently, we have engineered a new type of unimolecular micelle conjugated with anti-EGFR nanobody (Nb) exhibiting excellent tumor accumulation in a chemoresistant triple negative breast cancer patient derived xenograft (PDX) mouse model.

Keywords: Image classification, Image feature extraction, X-ray imaging applications
Abstract: To compare the performances of two kinds of computer-aided analysis algorithms of diffuse lung diseases on chest radiographs. One is a feature extraction based method using speeded up robust features (SURF) and a bag-of-features, and the other is a non-feature extraction based method using convolutional neural network (CNN).

Keywords: Image classification, Image segmentation, X-ray imaging applications
Abstract: This minisymposia presentation introduces image analysis methods for obtaining characteristic information from dental x-ray images for dental forensic identification. Unlike previous methods, our proposed methods using deep convolutional neural networks can classify tooth types and tooth conditions without precise segmentation of individual teeth. Such systems have a potential for assisting in systematic filing of antemortem dental information and reducing psychological burden in postmortem examination in large scale disasters.

Keywords: Image feature extraction
Abstract: Structure primitives and clustering are used to automatically learn prevalent structure patterns in complex and irregular shaped objects. The method is inspired by textons for texture analysis and is used here to characterize marbling patterns in beef.

Keywords: Image segmentation, X-ray CT imaging
Abstract: Abstract— This paper aims to describe an emerging method, the 3D statistical region merging method in medical image analysis. The method is a 3D generalization of the Statistical Region Merging method proposed by Nock and Nielsen in 2004. The 3D generalized statistical region merging method take into consideration of the group assignments of 3D neighboring voxels in a sequence of images such as CT images. The superiority of the 3D statistical region merging method is demonstrated in abdominal CT segmentation in this paper.

Keywords: Cardiovascular and respiratory system modeling - Gas exchange models, Pulmonary and critical care - Bioengineering applications in Intensive care, Cardiovascular and respiratory system modeling - Cardiovascular-Respiratory Interactions
Abstract: Deadspace gas mixing with stimulation of respiratory ventilation through cardiogenic oscillations is an important physiological mechanism during apnea. In this study, a novel method is proposed to investigate the effect of these mechanisms in silico. We extended an existing computational model of cardio-pulmonary physiology to include the apneic state, cardiogenic oscillations, gas mixing within the anatomical deadspace and insufflation into the trachea. The new model was validated against data published in an experimental animal study. Computational simulations confirmed that the model accurately reproduce the experimental data. This new model can be used to investigate the physiological mechanisms underlying clearance of carbon dioxide during apnea, and hence to develop more effective ventilation strategies for apneic patients.

Keywords: Pulmonary and critical care - Bioengineering applications in Intensive care, Pulmonary and critical care - Pulmonary disease, Pulmonary and critical care - Ventilatory Assist Devices
Abstract: Typical modern mechanical ventilators offer the clinician the possibility of automatically performing end-inspiratory occlusion maneuvers. The static conditions induced by such maneuvers are indeed favorable to estimating patient's respiratory mechanics in terms of total resistance (R) and elastance (E). These are parameters of wide clinical interest. However, in the presence of intrinsic PEEP, the standard formula used to compute E via the occlusion maneuver is known to be inaccurate. In this paper we propose an alternative method for the estimation of E via the occlusion maneuver that eliminates the bias by leveraging concepts derived from physiological modeling of the respiratory system dynamics. The proposed method is also capable of accounting for respiratory efforts triggering the breath, and hence can be applied in both passive and spontaneously breathing conditions.

Keywords: Pulmonary and critical care - Ventilatory Assist Devices, Pulmonary and critical care - Pulmonary disease
Abstract: In this paper we present an algorithm for the estimation of leaks in non-invasive ventilation. Accurate estimation of leaks is a key component of a ventilator, since it determines the ventilator performance in terms of patient-ventilator synchrony and air volume delivery. In particular, in non-invasive ventilation, the patient flow is significantly different from the flow measured at the ventilator outlet. This is mostly due to the vent orifice along the tube that is used for exhalation, but also to the non-intentional leaks that occur elsewhere in the circuit (e.g., at the mask). Such leaks are traditionally quantified via a model with two parameters, but only one of them is continually updated – the other is fixed. The new algorithm allows for breath-by-breath update of both parameters. This was made possible by leveraging a model describing the patient respiratory mechanics.

Keywords: Cardiovascular and respiratory system modeling - Compartmental modeling, Cardiovascular and respiratory system modeling - Cardiovascular-Respiratory Interactions, Cardiovascular and respiratory system modeling - Gas exchange models
Abstract: We present the adaptation of an advanced cardiorespiratory model of acute respiratory distress syndrome in adult patients to pediatric pathophysiology. We describe how model equations and parameters were revised to represent the physiological characteristics of pediatric Acute Respiratory Distress Syndrome (ARDS) patients. The adapted model was matched to data from twelve mechanically ventilated patients diagnosed with Pediatric Acute Respiratory Distress Syndrome (PARDS), and was shown to reproduce the available clinical data accurately for all patients. This new model constitutes the first detailed computational simulator specifically tailored to PARDS patients, and can be used as an investigational tool for developing and evaluating novel therapeutic strategies.

Keywords: Pulmonary and critical care - Bioengineering applications in Intensive care, Systems modeling - Decision making, Cardiovascular and respiratory system modeling - Respiratory Control models
Abstract: Providing the correct level of pressure support ventilation depends upon the patients’ ability to breathe spontaneously. Spontaneous breathing depends both on respiratory muscle function, but also on the chemical drive to breathe. Chemical drive is often reduced due to the action of anaesthesia, meaning that correct ventilator settings and anaesthesia strategies are coupled, and can be described through mathematical models of respiratory drive. This presentation will describe these models, and ongoing research toward understanding the coupling between anaesthesia and respiration in support ventilation.

Keywords: Pulmonary and critical care - Pulmonary disease, Pulmonary and critical care - Pulmonary function testing & instrumentation, Cardiovascular and respiratory system modeling - Compartmental modeling
Abstract: Time-based capnography measures the partial pressure of carbon dioxide in exhaled air as a function of time. It is the standard of care in a variety of patient settings in the United States and Canada, including emergency care, perioperative care, and even in ambulance settings. In current clinical practice, the resultant waveform signal, or capnogram, is used merely to establish the presence of respiration or to determine the respiratory rate or end-tidal CO2 concentration. However, the capnogram shape also has diagnostic value, but is only assessed qualitatively, by visual inspection. Here, we review mechanistic and data-driven model-based approaches to quantitative analysis of the capnogram to aid diagnosis and screening in a variety of patient conditions.

Keywords: Muscle stimulation
Abstract: Numerical simulations can give better insight into current conduction inside the tissues and can lead to new or improved usage of the bioimpedance measurement method. For this purpose an understanding of basic mathematical concepts is necessary in particular knowledge of the differential equations derived from Maxwells’ equations describing electromagnetic phenomena together with their limitations. Electrical properties of tissues are strongly frequency dependent and well covered in the relevant literature. Several computer tools can facilitate numerical simulations of bioimpedance ranging from professional to free and open-source ones.

Keywords: Diagnostic devices - Physiological monitoring, Ambulatory diagnostic and therapeutic devices - Ambulatory and ADL technologies, Ambulatory diagnostic devices - Wellness monitoring technologies
Abstract: Bioimpedance measurement is a technique known to provide information on biological material properties, but the measurement is often clouded with other types of information. This paper discusses the relevant and irrelevant information in the measurement, and how to discern between them.

Keywords: Bio-electric sensors - Sensor systems, Physiological monitoring - Modeling and analysis, Physiological monitoring - Novel methods
Abstract: In order to test and calibrate an EIT (Electrical Impedance Tomography) system, many researchers rely on phantoms mimicking breast tissues. These phantoms are usually made of saline solutions, agar and/or vegetables, allowing the user to set the conductivity of the material by changing the salt concentration. Due to that fact that the dispersion behavior in the vicinity of megahertz is fundamental to detect carcinoma, this work aims to propose a phantom composed by a mixture of agar and gelatin (emulating the normal tissue) and a piece of carrot (emulating the carcinoma). It also investigates the frequency dependence from 1 to 10,000 kHz. The proposed phantom showed capability to mimic some absolute and relative electrical parameters used to detect cancer according to the literature. The differences in the impedance modulus were found to be more dispersive in the mimic carcinoma tissue type than the normal mimic one. It is simple to prepare, low cost, has similar electrical properties to the ones that have been used in the literature, better mechanical properties and longer life time. It can be concluded that gelatin-agar gel may have a high potential to be used as a breast tissue phantom.

Keywords: Diagnostic devices - Physiological monitoring, Health technology - Verification and validation, Clinical engineering
Abstract: Over the last decades the electrical bioimpedance technique (EBI) is been largely applied as a promising tool for tissue diagnosis and in-vivo biosignals monitoring. EBI resolution is still to be improved by new advances in the circuitry and modelling methods toward a reliable technique. As a result, it will decrease the costs and increase its applicability among the professionals in the medical area.

Keywords: Signal pattern classification, Data mining and processing in biosignals, Connectivity measurements
Abstract: As brain-computer interface technology continues to advance toward practical applications, it can be extended to decoding and synthesizing music from mental imagery. Such a brain-actuated music synthesizer is envisioned to serve as a music composition and performance tool for musicians and non-musicians, as well as a potential communication and rehabilitation tool for the disabled. In order to decode brain activity during music listening, let alone generate new compositions directly from brain activity, a much more detailed knowledge of neural activations and pathways must be developed. To accomplish this, brain recordings with sufficient spatio-temporal resolution must be examined. This work presents preliminary results of signal characterization of electrocorticographic (ECoG) activity during perception of simple musical rhythms.

Keywords: Connectivity measurements, Partial and total coherence
Abstract: We applied graph theory analyses to neuronal data from motor cortices of monkey. The result showed that existence of community structures and their differences in tuning properties for arm movements. This result suggests that brain computer interface (BCI) algorithms to incorporate network properties might improve performance of BCI system.

Keywords: Neural networks and support vector machines in biosignal processing and classification
Abstract: The present study aims to develop models to decode spatio-temporal neuronal ensemble firing activities in anterior lateral motor (ALM) cortical neurons into one of the two licking movements (left vs. right). Using the linear discriminant analysis (LDA) classifier and the neuronal selection scheme, we could decode movement direction from ALM neural activity during the delay period after a tactile cue as well as during the movement period with accuracy of as high as 85.46%.

Keywords: Time-frequency and time-scale analysis - Nonstationary processing, Adaptive filtering, Nonlinear dynamic analysis - Biomedical signals
Abstract: Neuronal tuning property could change abruptly in brain machine interface (BMI), which may result in the decay of the decoding performance. We propose a novel adaptive algorithm to capture the abrupt alteration of neuronal tuning preference from point process observation. Tested on synthetic neural data, the proposed adaptive algorithm succeeds in detecting the abrupt change in neuronal tuning, and contributes to a better reconstruction of kinematics than static model.

Keywords: Image classification, Multivariate image analysis, Brain image analysis
Abstract: Many children born preterm exhibit frontal executive dysfunction, behavioral problems including attentional deficit hyperactivity disorder and attention related learning disabilities. Anomalies in regional specificity of cortico-striato-thalamo-cortical circuits may underlie deficits in these disorders. Here, we applied a recently developed multivariate tensor-based morphometry (mTBM) method to extract features from the surface anatomy of the striatum (putamen and globus pallidus) between 17 preterm and 19 term-born neonates scanned at term-equivalent age. For such surface-based features, the feature dimension is usually much larger than the number of subjects. We used dictionary learning and sparse coding to effectively reduce the feature dimensions. With the new features, an Adaboost classifier was employed for binary group classification. The new framework outperformed several standard imaging measures in classification, and achieved 86% accuracy, 83% sensitivity and 89% specificity. The new approach combines the efficiency of sparse coding with the sensitivity of surface mTBM, and boosts classification performance.

Keywords: 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 anterior part of the right putamen between verbal scores and surface metric were revealed in the 18 and 21 age groups.

Keywords: Brain image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: Mapping out the development of the brain in early childhood is a critical part of understanding neurological disorders. The brain grows rapidly in early life, reaching 95% of the final volume by age 6. A normative atlas containing structural parameters that indicate development would be a powerful tool in understanding the progression of neurological diseases. Although some studies have begun exploring cortical development in pediatric imaging, sulci have not been examined extensively. Here, we study the changes in the Central Sulcus (CS), which is one of the earliest sulci to develop from the fetal stage, at early developmental age 1-3 years old using high resolution magnetic resonance images. Parameterization of the central sulcus was performed and results show us that the CS change corresponds to the development of the mouth and tongue regions.

Keywords: Planning and execution in surgical robotics, Surgical robotics, New technologies and methodologies in medical robotics
Abstract: Percutaneous techniques and robot-assisted surgical systems have enabled minimally invasive procedures that offer reduced scarring, recovery time, and complications compared to traditional open surgeries. Despite these improvements, access to diseased sites using the standard, straight needle-based percutaneous techniques is still limited for certain procedures due to intervening tissues. These limitations can be further exacerbated in specific patient groups, particularly pediatric patients, whose anatomy does not fit the traditional tools and systems. We therefore propose a patient-specific paradigm to design and fabricate dexterous, robotic tools based on the patient's preoperative images. In this paper, we present the main steps of our proposed paradigm -- image-based path planning, robot design, and fabrication -- along with an example case that focuses on a class of dexterous, snake-like tools called concentric tube robots. We demonstrate planning a safe path using a patient's preoperative ultrasound images. We then determine the concentric tube robot parameters needed to achieve this path, and finally, we use 3-D printing to fabricate the patient-specific robot.

Keywords: Ultrasound imaging - Other organs, Brain image analysis, Image feature extraction
Abstract: Premature neonates with intraventricular hemorrhage (IVH) followed by post hemorrhagic hydrocephalus (PHH) are at high risk for brain injury. Cranial ultrasound (CUS) is used for monitoring of premature neonates during the first weeks after birth to identify IVH and follow the progression to PHH. However, the lack of a standardized method for CUS evaluation has led to significant variability in decision making regarding treatment. We propose a quantitative imaging tool for the evaluation of PHH on CUS for premature neonates based on morphological features of the cerebral ventricles. We retrospectively studied 64 extremely premature neonates born less than 29 weeks gestational age, less than 1,500 grams weight at birth, admitted to our center within two weeks of life, and diagnosed with different grades of IVH. We extracted morphological features of the lateral ventricles from CUS imaging using image analysis techniques to compare neonates who needed a temporizing intervention to treat PHH to the ones who did not. From the original set of features, an optimal ranking was obtained based on linear support vector machine. A subset of features was subsequently selected that maximizes the overall accuracy level. Regarding whether or not there was a need for temporizing intervention, we predicted the outcome of PHH with an improved accuracy level of 84%, compared to the 76% rate obtained by linear manual measurement. The proposed imaging tool allowed us to establish a quantitative method for PHH evaluation on CUS in extremely premature neonates with IVH. Further studies will help standardize the evaluation of CUS in those neonates to institute treatments earlier and improve outcomes.

Keywords: Optical imaging - Coherence tomography, Ultrasound imaging - Vascular imaging, Optical imaging and microscopy - Near infra-red spectroscopy
Abstract: A large majority of patients presenting with acute coronary syndrome is treated by stent implantation, in a procedure called percutaneous coronary intervention (PCI). Such interventions have a high acute success rate, but see a persistent 10-15% repeat revascularization rate due to stent failure or symptoms arising from different sites in the coronary circulation. Intravascular imaging guidance brings a marked improvement in outcome measures (including reduced reintervention). In Europe and the US, however, intravascular imaging is used in only a small fraction of PCIs, due to factors like cost, time, increased X-ray and contrast use, and the required operator expertise. In this overview, I will discuss how these perceived limitations of catheter-based interventional imaging can be ameliorated or overcome with technical improvements to the imaging technologies we apply. Highlighting developments in intravascular optical coherence tomography and photoacoustic imaging, I will focus on how advanced acquisition strategies can improve imaging speed and image contrast, with the aim to make intravascular imaging easier to use and interpret.

Keywords: Multimodal imaging, Ultrasound imaging - Vascular imaging, Optical imaging and microscopy - Fluorescence microscopy
Abstract: In this communication we report the design and testing of a intravascular catheter system able to simultaneously acquire co-registered structural and biochemical images of coronaries in vivo without the need of exogenous contrast agents. The system combines fluorescence lifetime imaging (FLIm) and intravascular ultrasound (IVUS) capabilities into a 3.7 Fr rotational catheter.

Keywords: Multimodal molecular imaging, Optical molecular imaging, Cardiac imaging and image analysis
Abstract: Cardiovascular imaging in recent years has focused on the biological aspects of coronary atherosclerosis, aiming to evaluate its natural course of the plaque and to predict future coronary events. Multi-modal biological imaging combined with intravascular OCT, albeit still in its early stage, is expected to shed a light in the precise diagnosis of coronary atherosclerosis.

Keywords: Integrated wearable and portable systems, Portable miniaturized systems, Wearable body-compliant, flexible and printed electronics
Abstract: In this paper, we demonstrate that a low power flash FPGA based micro-system can provide a low power programmable interface for closed loop brain implant interface.The proposed micro-system receives recording local field potential (LFP) signals from an implanted probe, performs closed loop control using a first order control system, then converts the signal into a stimulus patter. Stimulus is achieved through optoelectronic optogenetic means on the probe. The long-term target is for both fundamental neuroscience applications and for clinical use in treating epilepsy. Utilizing our device the closed loop processing consumes only 14nJ of power per PID cycle compared to 1.52J per cycle for a micro-controller implementation which is comparable to the power consumption of the optoelectrode probe.

Keywords: Integrated sensor systems, Wearable body-compliant, flexible and printed electronics, Physiological monitoring - Novel methods
Abstract: To improve the unobtrusiveness and comfortableness of sleep monitoring, we proposed the design and implementation of a smart mat which utilized flexible pressure sensor array and printed electrodes to monitor physiological and behavioral data during sleep. With the proposed novel soft seven-layer structure, the smart mat system can measure pressure distribution images and calculate respiratory rate of subjects. The function of this system was realized with pressure acquisition circuit controlled by Arduino processor. Experiments for pressure distribution tests and respiratory rate measurements were carried out to validate the performance of the proposed system. The experimental results demonstrated that the smart mat could be a viable option for unobtrusive sleep monitoring in the future.

Keywords: Chemo/bio-sensing - Biological sensors and systems, Portable miniaturized systems
Abstract: Among all infections, urinary tract infections (UTI) are one of the most common. Nowadays the procedures to analyze urine and consequently detect UTI are often painful and time-consuming. Recent studies about microfluidic paper based devices have developed the interest of researchers due their outstanding characteristics. In this paper is presented a novel design for a microfluidic paper-based device for screening and analysis of multiple biomarkers from urine samples on diapers. The device consists on a set of eight layers. It was designed based on the previous attempts to improve and overcome some problems detected as the continuous entrance of fluids, the possibility of contamination and the invalidity of results due to communication between different reagent pads. One approach was create a “self-locking” mechanism that closes the sample inlet in approximately four minutes solving the first two problems. Furthermore, is important that comfort is guaranteed, hence a device with a total thickness of 5,3 mm is presented. This device can keep the results for eight hours and can be used as a low-cost and more effective alternative than conventional methods being a strategy with potential for the diagnostic and analysis of biological samples in the future improving healthcare

Keywords: Chemo/bio-sensing - Biological sensors and systems, Integrated wearable and portable systems
Abstract: Urinalysis allows obtaining information about many of the body’s functions. Normally, urine is a very easily collected specimen, but in some cases, it can be an uncomfortable process. Moreover, standard methods such as microscopic examination or urine culturing represent time-consuming procedures. In this regard, this work explores the design of a device that combines disposable diapers, as a platform for sample collection, with part of the dipsticks method. Even though this is a desirable combination for a faster and comfortable approach, it is also convenient to extend the prescribed time, provided by the dipstick manufacturer, to read the results. This allows the stability and validity of the results for longer, preferably until the moment of the diaper changing, and it was also considered in the design. The desired conditions were reproduced for the reagent pads of glucose and nitrite. The results show that the RGB profile, obtained in glucose detection, varied after approximately 60 min. However, when the developed colors are visually compared with the reference colors, the results suggest that there were not significant variations. In the nitrite assay, by analyzing the RGB profile or by visual comparison with the reference colors, the color does not appear to change over time, but the results were not the expected. It seems that the proposed design is promising regarding results stability, but more experiments need to be done to check if the variations are constant for more biomarkers in the conditions provided by the device.

Keywords: Chemo/bio-sensing - Chemical sensors and systems, Wearable low power, wireless sensing methods, Portable miniaturized systems
Abstract: This paper describes the development of an array of individually addressable pH sensitive microneedles using injection moulding and their integration within a portable device for real-time wireless recording of pH distributions in biological samples. The fabricated microneedles are subjected to gold pat- terning followed by electrodeposition of iridium oxide to sensitize them to 0.07 units of pH change. Miniaturised electronics suitable for the sensors readout, analog-to-digital conversion and wireless transmission of the potentiometric data are embodied within the device, enabling it to measure real-time pH of soft biological samples such as muscles. In this paper, real-time recording of the cardiac pH distribution, during ischemia followed by reperfusion cycles in cardiac muscles of male Wistar rats has been demonstrated by using the microneedle array.

Keywords: Integrated wearable and portable systems, Integrated sensor systems, Implantable sensors
Abstract: In this paper, we discuss the structure and characteristics of carbon nanospikes. We also compare carbon nanospikes with previously grown carbon nanostructures known as vertically aligned carbon nanofibers (VACNF). Plasma enhanced chemical vapor deposition (PECVD) is used to fabricate both the nanospikes and VACNF. However, carbon nanospikes do not require a catalyst for the growth process, whereas VACNF requires a catalyst in the growth process. This facilitates batch fabrication with greater reproducibility. Scanning electron microscope images and Raman spectroscopy show that carbon nanospikes fabricated on silver wires will show superior performance.

Keywords:
Abstract:

Keywords: Biomaterial-cell interactions - Functional biomaterials, Micro- and nano-technology, Translational issues in tissue engineering and biomaterials
Abstract:

Keywords: Biomaterial-cell interactions - Functional biomaterials, Biomaterials - Sensors for evaluation of drug activity
Abstract:

Keywords: Stem cells - Engineered matrices for stem cells, Stem cells - Tissue morphogenesis
Abstract:

Keywords: Scaffolds in tissue engineering - Biofabrication, Scaffolds in tissue engineering - Patterned 3D
Abstract:

Keywords: Motor neuroprostheses, Brain-computer/machine interface, Brain physiology and modeling - Neural dynamics and computation
Abstract: Recordings from invasive implants can degrade over time, resulting in a loss of spiking activity for some electrodes. For brain-machine interfaces (BMI), such a signal degradation lowers control performance. Achieving reliable performance over time is critical for BMI clinical viability. One approach to improve BMI longevity is to simultaneously use spikes and other recording modalities such as local field potentials (LFP), which are more robust to signal degradation over time. We have developed a multiscale decoder that can simultaneously model the different statistical profiles of multiscale spike/LFP activity (discrete spikes vs. continuous LFP). This decoder can also run at multiple time-scales (millisecond for spikes vs. tens of milliseconds for LFP). Here, we validate the multiscale decoder for estimating the movement of 7 major upper-arm joint angles in a non-human primate (NHP) during a 3D reach-to-grasp task. The multiscale decoder uses motor cortical spike/LFP recordings as its input. We show that the multiscale decoder can improve decoding accuracy by adding information from LFP to spikes, while running at the fast millisecond time-scale of the spiking activity. Moreover, this improvement is achieved using relatively few LFP channels, demonstrating the robustness of the approach. These results suggest that using multiscale decoders has the potential to improve the reliability and longevity of BMIs.

Keywords: Motor neuroprostheses, Brain-computer/machine interface, Brain physiology and modeling - Neural dynamics and computation
Abstract: Technological advances have enabled the simultaneous recording of multiscale neural activity consisting of spikes, local field potential (LFP), and electrocorticogram (ECoG). Developing models that describe the encoding of behavior within multiscale activity is essential both for understanding neural mechanisms and for various neurotechnologies such as brain-machine interfaces (BMI). Multiscale recordings consist of signals with different statistical profiles and time-scales. While encoding models have been developed for each scale of activity alone, developing statistical models that simultaneously characterize discrete spike and continuous LFP/ECoG recordings and their various time-scales is a major challenge. To address this challenge, we have recently proposed a multiscale state-space encoding model for combined spike/LFP/ECoG recordings. However, methods to learn these state-space models from data are still lacking. Here, we develop an unsupervised learning algorithm for multiscale state-space models. Given a multiscale dataset, our algorithm finds the maximum-likelihood estimate of the state-space model parameters using a new expectation-maximization (EM) technique. We show that the new algorithm can learn the encoding model accurately from simulated multiscale data. We also show that the learned model can be used to decode arm movement trajectories from simulated multiscale activity. These multiscale models have the potential to improve the performance and robustness of various neurotechnologies.

Keywords: Motor neuroprostheses - Neuromuscular stimulation
Abstract: The voluntary participation of the paralyzed patients is crucial for the functional electrical stimulation (FES) therapy. In this study, we developed a strategy called “EMG Bridge” (EMGB) for volitional control of multiple movements using FES technique. The surface electromyography (sEMG) signals of the agonist muscles were transformed to stimulation pulses with various pulse width and frequency to stimulate the target paralyzed muscles using MAV/NSS co-modulation (MNDC) algorithm we proposed recently. Motion pattern classification based on linear discriminant analysis (LDA) was included to recognize the motion status and mapping the sEMG detection channel to the corresponding stimulation channel. A prototype EMGB system was built for real-time control of four hand movements. The test results showed that the movements can be reproduced with a successful rate of 92.5±3.5%. The angle trajectory of wrist joint and metacarpal-phalangeal joint can be mimicked with a maximum cross-correlation coefficient > 0.84 and a latency less than 300 ms.

Keywords: Motor neuroprostheses - Neuromuscular stimulation, Neurorehabilitation, Neural interfaces - Regeneration
Abstract: Low-intensity ultrasound can be used for noninvasive neuromuscular stimulation to enhance the recovery of injured peripheral nerves and muscles. In order to improve the stimulation efficiency, focused ultrasound (FUS) with a high energy density should be used. Because the FUS is concentrated at a miniaturized focal area, deliberate targeting with focal depth control is required to achieve the accurate stimulation. In this study, a low-intensity focused ultrasound (LIFU) stimulator with a focal depth controller is presented as an effective ultrasound stimulation system for the neuromuscular rehabilitation. In the developed stimulator, the target depth is adjusted by the focal depth controller with the brief structure. The experimental results show that the developed stimulator can generate the high quality LIFU stimulation with the spatial average-pulse average intensity (ISAPA) of 496 mW/cm2, which is sufficient to improve the rehabilitation of injured peripheral nerves and muscles. In the in vivo experiment, a rat’s sciatic nerve was successfully targeted by the developed stimulator. These results indicate that the developed stimulator can be used to improve the quality of the ultrasound stimulation for neuromuscular rehabilitation by targeting a specific area.

Keywords: Motor neuroprostheses - Prostheses, Motor neuroprostheses - Robotics, Brain-computer/machine interface
Abstract: Numerous hand exoskeletons have been proposed in the literature with the aim of assisting or rehabilitating victims of stroke, brain/spinal cord injury, or other causes of hand paralysis. In this paper a new 3D printed soft hand exoskeleton, HEXOES (Hand Exoskeleton with Embedded Synergies), is introduced and mechanically characterized. MCP and PIP joints had measured maximum flexion angles of 53.7 ± 16.9° and 39.9 ± 13.4°, respectively; and maximum MCP and PIP angular velocities of 94.5 ± 41.9 degrees/s and 74.6 ± 67.3 degrees/s, respectively. These estimates indicate that the mechanical design has range of motion and angular velocity characteristics that meet the requirements for synergy-based control. When coupled with the proposed control loop, HEXOES can be used in the future as a test-bed for synergy-based clinical hand rehabilitation.

Keywords: Neural interfaces - Bioelectric sensors, Neural interfaces - Implantable systems, Neural interfaces - Body interfaces
Abstract: This paper describes an integrated multichannel neural recording analog front end (AFE) with a novel area-efficient driven right leg (DRL) circuit to improve the system common mode rejection ratio (CMRR). The proposed AFE consists of an AC-coupled low-noise programmable-gain amplifier, an area-efficient DRL block and a 10-bit SAR ADC. Compared to conventional DRL circuit, the proposed capacitor-less DRL design consumes 90% less chip area with comparable CMRR performance, making it ideal for multichannel biomedical recording applications. The AFE circuit has been designed in a standard 0.18-µm CMOS process. Post-layout simulation results show that the AFE provides two gain settings of 54dB/60dB while consuming 1 µA per channel under a supply voltage of 1 V. The input-referred noise of the CCIA amplifier integrated from 1 Hz to 10k Hz is only 4 µVrms and the system CMRR is 110 dB.

Keywords: Neural interfaces - Bioelectric sensors, Neural interfaces - Implantable systems, Brain-computer/machine interface
Abstract: Advances in minimally-invasive, distributed biological interface nodes enable possibilities for networks of sensors and actuators to connect the brain with external devices. The recent development of the neural dust sensor mote has shown that utilizing ultrasound backscatter communication enables untethered sub-mm neural recording devices. These implanted sensor motes require a wearable external ultrasound interrogation device to enable in-vivo, freely-behaving neural interface experiments. However, minimizing the complexity and size of the implanted sensors shifts the power and processing burden to the external interrogator. In this paper, we present an ultrasound backscatter interrogator that supports realtime backscatter processing in a rodent-wearable, completely wireless device. We demonstrate a generic digital encoding scheme which is intended for transmitting neural information. The system integrates a front-end ultrasonic interface ASIC with off-the-shelf components to enable a highly compact ultrasound interrogation device intended for rodent neural interface experiments but applicable to other model systems.

Keywords: Neural interfaces - Implantable systems, Neural interfaces - Tissue-electrode interface, Motor neuroprostheses - Prostheses
Abstract: This paper proposes a compact handheld device for magnetically inserting a neural interface into a peripheral nervous system (PNS). Users can pull and hold a flexible peripheral nerve (e.g., sciatic nerve) at the front of the device for the accurate and stable targeting process. The device automatically inserts a neural interface using magnetic impacts that are generated by a miniature motor and a pair of magnets. We investigate the characteristics of the employed mechanism, and present the preliminary experimental results.

Keywords: Neural interfaces - Implantable systems, Sensory neuroprostheses - Somatosensory, Neuromuscular systems - Computational modeling
Abstract: Providing realistic sensory feedback for prosthetic devices strongly relies on an accurate modelling of machine-nerve interfaces. Models of these interfaces in the peripheral nervous system usually neglect the effects that ephaptic coupling can have on the selectivity of stimulating electrodes. In this contribution, we study the ephaptic stimulation between myelinated axons and show its relation with the separation between fibers and the conductivity of the medium that surrounds them.

Keywords: Neural interfaces - Implantable systems, Neural interfaces - Microelectrode technology, Neural interfaces - Biomaterials
Abstract: This work proposes solutions to the current bulky packaged neural implants. We describe the next generation of miniaturized wirelessly powered neural interface that are distributed and free floating in the nervous system. This paper focuses on the microassembly, hermetic packaging and its effect on the inductive power link.

Keywords: Neural interfaces - Implantable systems, Neural interfaces - Microelectrode technology, Brain-computer/machine interface
Abstract: Detachable high-channel electrical connections pose a bottleneck on the path to active implants with higher numbers of electrode contacts and miniaturized geometries. Not only low-resistance, reproducible and reliable contacts have to be realized but also seals that ensure electrical insulation in the harsh body environment. Using planar contact arrangements one can resort to laser microprocessing leading to minimal size connectors. However, this poses the need to carefully design the topographies within the contact zone. In this study, we assess different methods to design the topography of planar contact pad arrays. Using topographical analysis and evaluation of electrical functionality, influential mechanisms were identified and two functional ones have been selected.

Keywords: Health Informatics - eHealth, Health Informatics - Mobile health, Health Informatics - Personal health records
Abstract: It is known that voice changes in various diseases, and attempts to detect disease by voice analysis have been performed since long ago. We developed a pathophysiological voice analysis technique, based on the speech emotion recognition technology. We have accumulated various evidences in this technology and have reached almost practical stage in the mental healthcare field.

Keywords: Bioinformatics - Bioinformatics for health monitoring
Abstract: This study includes developing an algorithm to estimate a disease condition based on speech. First, we recorded the speech including the reading of short sentences by healthy subjects, major depression patients, and patients suffering from Parkinson's disease in a hospital consultation room. Next, the acoustic feature quantity was calculated based on the speech, and an algorithm for classifying healthy subjects and patients was developed. This algorithm was evaluated using the disordered voice database, containing information related to several articulatory disorders. Results indicated that the algorithm performed fairly well in classifying healthy subjects and patients, which suggested the usefulness of the classification algorithm in estimating disease conditions based on speech.

Keywords: Bioinformatics - Bioinformatics for health monitoring
Abstract: In this paper, we propose a novel voice indicator capable of distinguishing healthy subjects from patients with Parkinson’s disease. Specifically, we propose an indicator based on the relationship between two indices, the Zero Crossing Rate (ZCR) used in fields such as Voice Activity Detection (VAD), and the Hurst Exponent (HE) used in stock price analysis.

Keywords: Health Informatics - Healthcare modeling and simulation, General and theoretical informatics - Machine learning, General and theoretical informatics - Pattern recognition
Abstract: We explore the reliability of using quantifiable voice acoustic measures for pairwise classification between speech samples of suicidal predisposition, major depressive disorders (MDD) and remitted/healthy subjects. Linear combinations of spectral and timing features of voice was also shown to predict clinical ratings of depression severity measured using the Hamilton Depression Rating Scale (HAMD) and Beck Depression Inventory-II (BDI-II).

Keywords: Bioinformatics - Bioinformatics for health monitoring
Abstract: The purpose of this study is to verify the impact of a deterioration of the sound quality of voice by a telephone line on estimating Vitality as the extent of depressive tendency based on voice analysis using MIMOSYS. First, the voices of about 1,000 people recorded using a recorder were prepared. Next, each voice was coded and resampled in preparation for transmission over a phone line. Vitalities obtained by analyzing the voices before and after these processes were compared. The results showed high correlation between the Vitality after coding and Vitality before coding, revealing that using a telephone would be an effective way to obtain voices.

Keywords: General and theoretical informatics - Statistical data analysis, General and theoretical informatics - Deep learning and big data to knowledge, General and theoretical informatics - Algorithms
Abstract:

Keywords: Cardiovascular and respiratory signal processing - Pulse transit time
Abstract: There are several previously published approaches of measuring local pulse transit time (PTT). One of these approaches is to use two optical sensors based on photoplethysmography (PPG). However, little information about reproducibility in PPG based PTT measurement is available. Therefore, we performed a small sample size study (n = 5) to investigate quantitative criteria for reproducible PTT measurement. The inflection point as a characteristic feature of the pulse wave showed the most stabile results under varying conditions. Furthermore, we found that correlation between related pulse waves could be used as a threshold for signal quality. We suggest to implement a real-time operator feedback based on the found criteria to ensure reproducible PTT measurements.

Keywords: Vascular mechanics and hemodynamics - Pulse wave velocity, Vascular mechanics and hemodynamics - Vascular Hemodynamics, Pulmonary and critical care - Pulmonary function testing & instrumentation
Abstract: In this study, pulse arrival time (PAT) was measured using a simple measurement setup consisting of arm electrocardiogram (ECG) and finger photoplethysmogram (PPG). Four alternative methods to calculate PAT from the measured signals were evaluated. PAT was calculated as the time delay between ECG R peak and one of the following points in the PPG waveform: peak (maximum value of PPG waveform), foot (minimum value of PPG waveform), dpeak (maximum value of the first derivative of PPG waveform) and ddpeak (maximum value of the second derivative of PPG waveform). In addition to PAT calculation, pulse period (PP) intervals based on the detected features were determined and compared to RR intervals derived from ECG signal. Based on the results obtained here, the most promising method to be used in PAT or PP calculation seems to be the dpeak detection method.

Keywords: Vascular mechanics and hemodynamics - Pulse wave velocity, Vascular mechanics and hemodynamics - Vascular mechanics, Pulmonary and critical care - Pulmonary function testing & instrumentation
Abstract: This study investigated the effect of coffee intake in pulse arrival time (PAT) and pulse wave velocity (PWV) measured with electrocardiogram (ECG) from arms and photoplethysmogram (PPG) from fingertip. 30 healthy participants were recruited to two measurement sessions, one arranged before and another one after the coffee intake. During each session, ECG and PPG were measured continuously for six minutes and PAT values calculated from ECG R-peak to the maximum of the first derivative of the PPG pulse. In addition, blood pressure (BP) was measured twice during each session with cuff based method. Coffee intake had statistically significant influence on both systolic and diastolic blood pressure, but not on PAT or PWV. Correlation between systolic blood pressure and PWV was 0.44. Individual calibration, additional derivatives of ECG and PPG such as heart rate, pulse pressure, or waveform characteristics could improve the correlation.

Keywords: Vascular mechanics and hemodynamics - Arterial pressure in cardiovascular disease, Vascular mechanics and hemodynamics - Pulse wave velocity, Cardiovascular regulation - Autonomic nervous system
Abstract: Renal denervation is a novel device based therapy promoted to reduce high blood pressure. We examined the impact of renal denervation on systolic blood pressure, renal function, and arterial stiffness in the Lewis Polycystic Kidney disease (LPK) rodent model of kidney disease. Animals were subjected to bilateral renal denervation or sham surgeries at age 6 and 12 weeks. Systolic blood pressure was monitored by tail-cuff plethysmography and renal function by urinalysis and creatinine clearance. At age 16 weeks, beat-to-beat aortic pulse wave velocity as a functional indicator of arterial stiffness was determined. Renal denervation produced an overall reduction in blood pressure in the LPK [(denervated 164±4 vs. sham-operated 180±6 mmHg, n = 6 per group, P=0.003)] and delayed, but did not prevent, the decline in renal function. Aortic pulse wave velocity was markedly elevated in the LPK compared with Lewis and was not altered by renal denervation in the LPK however a reduction was seen in the control Lewis animals. These results support the hypothesis that renal nerves contribute to secondary hypertension in conditions such as kidney disease.

Keywords: Vascular mechanics and hemodynamics - Vascular Disease, Vascular mechanics and hemodynamics - Vascular mechanics, Vascular mechanics and hemodynamics - Vascular Hemodynamics
Abstract: Central and peripheral arteries stiffening prominently affect hemodynamics thus increasing the risk of coronary heart disease, chronic kidney disease and end stage renal disease. There are several commercially available non-invasive measurement technologies for the evaluation of stiffness that are expensive, demand dedicated expertise and fall short for mass screening. Considering this, we have developed ARTSENS, a highly compact and portable image-free ultrasound device for evaluation of arterial stiffness. The capability of the device to perform accurate measurements of carotid artery stiffness has been validated through extensive in-vivo studies. In this paper we demonstrate the feasibility of using ARTSENS for measuring brachial artery stiffness. An inter-operator repeatability study was done based on in-vivo experiments on 9 young healthy subjects. The study included measurement of distension, end diastolic lumen diameter, arterial compliance and stiffness index performed both on carotid artery and brachial artery by two operators successively. The degree of agreement between the measurements made by operators has been investigated based on Bland-Altman plots and paired t-test. The measurements were populated within the limits of agreement. No statistically significant difference (p-values from paired t-test for end-diastolic diameter, distension, stiffness index, arterial compliance were 0.36, 0.24, 0.47 and 0.11 respectively) was seen for the brachial artery measurements performed by the two operators. The correlation between the measurement made by the operators was highly significant (r=0.86, p-value=0.003).

Keywords: Vascular mechanics and hemodynamics - Vascular Hemodynamics, Cardiovascular and respiratory system modeling - Vascular mechanics and hemodynamics
Abstract: Introduction: Measurement of hemodynamic parameters constitutes an important tool in the management of patients with cirrhosis. In recent years, non-invasive measurements have gain attention, due to the potential complications associated to invasive procedures. Objective: To characterize the hemodynamic alterations of cirrhosis in young and adult subjects, through a three element windkessel (WK3e). Methods: Individuals were divided in three age groups: Young Healthy group (control, CG), Young Cirrhotic Group (YCG) and Adult Cirrhotic Group (ACG). A Finapres® Nova device was used to obtain cardiac output (CO), heart rate and arterial blood pressure. Systemic arterial compliance (C), peripheral resistance (R) and characteristic impedance (ZC) were also provided. Effective arterial elastance (Ea), left ventricular work (LVW), input impedance, efficiency and model cutoff frequency (WKCF) were assessed based on the provided data. Results: CO resulted to be higher in ACG than in CG and YGC. LVW, C and WKCF showed an increase, while R and Ea showed a decrease. However, this behavior was not observed in YCG. Conclusion: Cirrhosis was properly modeled in young and adult subjects in terms of non-invasive measurements and a WK3e.

Keywords: Ambulatory Diagnostic devices - Point of care technologies, Health technology - Verification and validation, Diagnostic devices - Physiological monitoring
Abstract: Support Vector Machine (SVM) is a common classifier used for efficient classification with high accuracy. SVM shows high accuracy for classifying melanoma (skin cancer) clinical images within computer-aided diagnosis systems used by skin cancer specialists to detect melanoma early and save lives. We aim to develop a medical low-cost handheld device that runs a real-time embedded SVM-based diagnosis system for use in primary care for early detection of melanoma. In this paper, an optimized SVM classifier is implemented onto a recent FPGA platform using the latest design methodology to be embedded into the proposed device for realizing online efficient melanoma detection on a single system on chip/device. The hardware implementation results demonstrate a high classification accuracy of 97.9% and a significant acceleration factor of 26 from equivalent software implementation on an embedded processor, with 34% of resources utilization and 2 watts for power consumption. Consequently, the implemented system meets crucial embedded systems constraints of high performance and low cost, resources utilization and power consumption, while achieving high classification accuracy.

Keywords: Ambulatory diagnostic and therapeutic devices - Ambulatory and ADL technologies
Abstract: Gaming systems have been proven to be able to improve physical and cognitive skills. Several studies have integrated games to be part of rehabilitation program to improve functional ability of human limbs in performing activities of daily living. This paper presents the development of an interactive electromyogram (EMG) based PONG game for foot dorsifexion and plantarflexion exercise. PONG game is selected because it is a classical game that can be easily comprehended and can be played by young and old people. Two surface EMG sensors were used in this work to acquire muscle activity of tibialis anterior muscle and gastrocnemius muscle. An experimental study was carried out to examine the viability of this sensor as input to play the game. The results were satisfactory. This study further strengthens the viability of using muscle activity as an input for gaming system and as a mean to monitor patient rehabilitation progress.

Keywords: Ambulatory diagnostic and therapeutic devices - Ambulatory and ADL technologies, Ambulatory diagnostic devices - Wellness monitoring technologies, Health technology management and assessment
Abstract: An infrastructure to record, detect and label the behavioral patterns of children with Autism Spectrum Disorder (ASD) has been developed. The system incorporates 2 different sensor platforms which are wearable and static. The wearable system is based on accelerometer which detects behavioral patterns of a subject, while the static sensors are microphones and cameras which captures the sounds, images and videos of the subjects within a room. The video also provides ground truth for wearable sensor data analysis. The system labels the segment of video data upon detection of the autistic behavior. That is, it stores the time of the video when the activities are detected. Time-Frequency methods are used to extract features and Hidden Markov Model (HMM) are used for analyzing the accelerometer signal. Using these methods, we are able to achieve 91.5% of classification rate for behavioral patterns studied in this paper which is used to label and save data.

Keywords: Ambulatory Therapeutic Devices - Personalized therapeutic devices and emergency response systems
Abstract: Epilepsy is a group of diseases caused by excessive neuronal activities, and one-quarter of the patients do not become seizure-free by the existing treatments. The potential treatments include focal brain cooling, which aims to cool the region where the excessive neuronal activities begin. We are developing a focal brain cooling system. The system delivers cold saline to a cranially implanted cooling device. The outflow is cooled by a Peltier device and pumped for circulation. The Peltier device and the pump are activated only when a seizure is predicted. In this research, the length of time for cooling the brain was calculated with a computational fluid dynamics (CFD)-based model of the focal brain cooling system. As a result, it takes less than 10 minutes for the average temperature 2 mm below the cooling device to reach 25.0 ◦C. It is much shorter than the time from seizure prediction to seizure onset when an existing algorithm for prediction is used.

Keywords: New sensing techniques, Mechanical sensors and systems, Integrated wearable and portable systems
Abstract: Interpersonal relationships are necessary for successful daily functioning and wellbeing. Numerous studies have demonstrated the importance of social connectivity for mental health, both through direct peer-to-peer influence and by the location of individuals within their social network. Passive monitoring using smartphones provides an advanced tool to map social networks based on the proximity between individuals. This study investigates the feasibility of using a smartphone app to measure and assess the relationship between social network metrics and mental health. The app collected Bluetooth and mental health data in 63 participants. Social networks of proximity were estimated from Bluetooth data and 95% of the edges were scanned every 30 minutes. The majority of participants found this method of data collection acceptable and reported that they would be likely to participate in future studies using this app. These findings demonstrate the feasibility of using a smartphone app that participants can install on their own phone to investigate the relationship between social connectivity and mental health.

Keywords: Medical technology - Design and development, Medical technology - Innovation
Abstract: Intraoperative radiation therapy (IORT) involves delivering high doses of radiation directly to tumors while sparing healthy tissues in a surgical setting. Current IORT systems are limited in their lack of image guidance and variable needs for shielded operating rooms. They also lack the capability to deliver non-uniform therapeutic radiation to irregular shaped clinical targets. We developed a scanning beam IORT system (SBIORT) to overcome these limitations. SBIORT consists of a low energy x-ray source, a custom compact dynamic x-ray collimator system, a robotic arm, and a 3D surface imaging module. Here we describe the design and validation of the compact dynamic x-ray collimator system and the 3D surface imaging module for use in SBIORT. The proposed collimator can achieve a leaf position accuracy of ± 0.25 mm (95% confidence interval). Phantom studies indicated the 3D surface-imaging module has an accuracy of 1.0 ± 0.6 mm with ability to obtain high resolution surface image within 5 seconds. SBIORT is a novel approach to deliver conformal intensity-modulated intraoperative radiation therapy.

Keywords: Ultrasound imaging - Other organs, Image segmentation, Image feature extraction
Abstract: Telesonography involves transmission of ultrasound video from remote areas to the doctors for getting diagnosis. Due to the lack of trained sonographers in remote areas, the ultrasound videos scanned by these untrained persons do not contain the proper information that is required by a physician. As compared to standard methods for video transmission, mHealth driven systems need to be developed for transmitting valid medical videos. To overcome this problem, we are proposing an organ validation algorithm to evaluate the ultrasound video based on the content present. This will guide the semi skilled person to acquire the representative data from patient. Advancement in smartphone technology allows us to perform high medical image processing on smartphone. In this paper we have developed an Application (APP) for a smartphone which can automatically detect the valid frames (which consist of clear organ visibility) in an ultrasound video and ignores the invalid frames (which consist of no-organ visibility), and produces a compressed sized video. This is done by extracting the GIST features from the Region of Interest (ROI) of the frame and then classifying the frame using SVM classifier with quadratic kernel. The developed application resulted with the accuracy of 94.93% in classifying valid and invalid images.

Keywords: Micro and Nano formulation - Nanotechnology/Nanoparticles
Abstract: Chemo-immunotherapy which combines chemotherapeutics with immune-modulating agents represents an appealing approach to improving cancer therapy [1, 2]. To maximize its efficacy, differential and precise targeting of the multiple therapeutics into corresponding cells is desirable. Here we develop an immunostimulatory nanocarrier that simultaneously loads platinum (Pt)-based chemotherapeutic prodrug and BLZ-945, a small molecule inhibitor of colony stimulating factor 1 receptor (CSF-1R) of tumor-associated macrophages (TAMs) [3], to spatially target tumor cells and TAMs for cancer chemo-immunotherapy. The nanoparticles (denoted as BLZ-945SCNs/Pt) undergo supersensitive structure collapse through responding to tumor acidity, along with instantaneous release of BLZ-945 and Pt-prodrug conjugated small particles. The extracellularly released BLZ-945 could be taken up by TAMs, which locate preferentially in the perivascular region [4], to directly disturb CSF-1/CSF-1R signaling pathway to suppress TAMs and modulate the tumor immune microenvironment, while the released small particles carrying Pt-prodrug could penetrate deeply into bulk tumor to kill more cancer cells [5], realizing synergistic antitumor effect of chemo- and immunotherapy. Our in vivo studies demonstrate that the co-delivery nanocarrier outperforms monotherapy in varying tumor models.

Keywords: Drug delivery routes - Transdermal drug delivery
Abstract: Controllable jet injection could provide a fast, repeatable method for delivery of small volume, high viscosity drugs to the dermis. We have developed a compact, lightweight jet injector that uses a small commercial rotary motor and differential screw to create a jet of fluid with sufficient force to deliver drug into tissue. Clear tissue analogue and ex vivo tissue were used to demonstrate repeatable delivery of <10 μL of fluid using the device. The fluid when forced through a narrow orifice (108 μm) reaches a velocity of >250 m/s with an injection time of <3 ms.

Keywords: Drug delivery routes - Transdermal drug delivery
Abstract: The effect of varying velocity during jet injection on the dispersion of fluid into tissue is investigated using a custom-built X-ray imaging system. Injections are performed into ex-vivo porcine abdominal tissue using a voice coil actuated injection device. Single velocity and two-phase velocity injections reveal the complex nature of the dispersion of the fluid jet in layered tissue and highlight the effects of changing the jet velocity following the initial penetration of the liquid into the tissue.

Keywords: Drug release and solubility - Controlled/Sustained/Modified release, Drug release and solubility - Dissolution/Mass transfer, Drug formulaion - Polymers
Abstract: An implantable drug delivery device with an inflatable/deflatable reservoir, a release rate controlling micro-channel, a syringe attachable inlet, a unidirectional channel preventing drug substances from flowing back into the inlet, and a magnetic polydimethylsiloxane (PDMS) outlet capable of on-demand drug release was designed and fabricated to realize effective drug therapy. The reservoir is refillable through the inlet, thus the device is reusable. The device does not employ photolithography patterning. Vapor coating, a precise cutting plotter, a vinyl adhesive material, and oxygen plasma treatment were employed to fabricate the device. Drug release rate was controlled by the micro-channel connected to the reservoir. Micro-channels with widths 300μm, 400μm, 500μm and depth 95μm were tested. Drug release is executed by deflecting the magnetic membrane outlet by an external electromagnetic field generating source, a permanent magnet or a planar copper coil.

Keywords: Drug delivery routes - Parenteral drug delivery, Drug delivery routes - Transdermal drug delivery, Drug delivery routes - Ocular drug delivery
Abstract: A rigid, compact, multi-piece ampoule has been designed and fabricated as a platform to conveniently house different nozzle inserts one at a time to offer increased jet consistency and functional flexibility. Three different nozzle geometries that respectively produce a single axial jet, radial jets, and intersecting jets are designed and fabricated. Through high-speed imaging and injection into tissue analog and ex vivo porcine tissue, these nozzles are demonstrated to have potential for a range of injection applications including intradermal, intratympanic, and traditional uses. The presented ampoule is more successful than a commercial jet injection ampoule in adhering to desired position and velocity trajectories and the energy efficiencies of jet ejections are compared.

Keywords: Micro and Nano formulation - Nanotechnology/Nanoparticles
Abstract: This paper describes the preparation and characterization of polymeric nanoparticles loaded with a potent anti-tumor metal chelator, Di-2-pyridylketone-4,4- dimethyl-3-thiosemicarbazone (Dp44mT) for delivery to cancer cells. Metal chelators have been increasingly studied for their anti-cancer properties that rely on the high demand of neoplastic cells for iron. Dp44mT has previously shown great antiproliferative characteristics in several cancers including breast cancer and melanoma. To further expand the application of this highly cytotoxic agent for cancer treatment and to enable its specific delivery to malignant cells, here we apply nano-scale particles (NPs) of biodegradable poly(lactic-co-glycolide) (PLGA) for encapsulation of Dp44mT and evaluate its effectiveness in vitro. The results demonstrated that Dp44mT was efficiently encapsulated in PLGA particles. Resulting NPs were uniform in size and shape and had good colloidal stability. Moreover, Dp44mT encapsulation in PLGA enhanced the water solubility of this agent. Lastly, the present formulation showed high level of cytotoxicity in glioma cells. Together, these results show the potential of PLGA NPs as a nano-carrier for Dp44mT with no apparent impact on the anti-tumor activity of this compound.

Keywords: Image segmentation, Deformable image registration, Magnetic resonance imaging - MR neuroimaging
Abstract: Accurate rodent brain extraction is one of the basic steps for many translational study using Magnetic Resonance Imaging (MRI). In this paper, we present a new approach to model the rodent brain variation using non-rigid B-spline image registration for the brain extraction in MRI images. We model the shape and appearance with the B-spline parameters together with a mean brain image. Followed by a method using multi-expert, we refine the brain extraction region. Compared with the image-based template model using cross-correlation, the performance for rodent brain extraction has shown much improvement on one data set while maintaining the similar yet more consistent performance for another. Both template based methods however outperform the voxel based method (3D PCNN) and a modified BET version for rodent brain extraction.

Keywords: Image segmentation, Deformable image registration, Magnetic resonance imaging - Other organs
Abstract: Manual and automated segmentation of individual muscles in magnetic resonance images have been recognized as challenging given the high variability of shapes between muscles and subjects and the discontinuity or lack of visible boundaries between muscles. In the present study, we proposed an original algorithm allowing a semi-automatic transversal propagation of manually-drawn masks. Our strategy was based on several ascending and descending non-linear registration approaches which is similar to the estimation of a Lagrangian trajectory applied to manual masks. Using several manually-segmented slices, we have evaluated our algorithm on the four muscles of the quadriceps femoris group. We mainly showed that our 3D propagated segmentation was very accurate with an averaged Dice similarity coefficient value higher than 0.91 for the minimal manual input of only two manually-segmented slices.

Keywords: Deformable image registration, Optical imaging - Confocal microscopy
Abstract: Precise three-dimensional mapping of a large number of gene expression patterns, neuronal types and connections to an anatomical reference helps us to understand the vertebrate brain and its development. Zebrafish has evolved as a model organism for such study. In this paper, we propose a novel non-rigid registration algorithm for volumetric zebrafish larval image datasets. A coarse affine registration using the LBFGS algorithm is applied first on the moving dataset. We then divide this coarsely registered moving image and the reference image into a union of overlapping patches. Minimum weight bipartite graph matching algorithm is employed to find the correspondence between the two sets of patches. The corresponding patches are then registered using the diffeomorphic demons method with proper intra-patch regularization. For each voxel lying in the overlapping regions, we impose inter-patch regularization through a composite transformation obtained from the adjacent transformation fields. Experimental results on four multi-view confocal 3D datasets show the advantage of the proposed solution over the existing ViBE-Z software.

Keywords: Deformable image registration, Magnetic resonance imaging - Contrast-enhanced dynamic MRI
Abstract: This study presents an accelerated implementation of a two-dimensional moving mesh point correspondence algorithm using a GPU for tracking mobile tumor boundaries during radiation therapy. Normal CPU implementation of this algorithm is computationally intensive and time-consuming which limits its clinical utility, hence the need for a faster GPU implementation. One of the computationally intensive parts of the registration algorithm involves numerically solving a partial differential equation. In this paper we demonstrate that the computational performance of the algorithms can be improved by utilizing a shared memory implementation on the GPU. Evaluations in comparison to 600 manually drawn contours showed that the proposed GPU-based tracking of the tumor boundaries yielded similar level of accuracy as the CPU based approach with improved computational efficiency.

Keywords: Image classification, Image feature extraction, Multivariate image analysis
Abstract: Computational analysis of cell dynamic morphology in time-lapse images has become a new topic of biomedical research. For single cell, it is a challenging task to consider the spatial inconsistency and the temporal accumulation of cell deformation. This paper introduces an innovative automate analysis method, in which temporal features of contour point deformation are captured and then local deformation pattern is modeled to characterize cell dynamic morphology and predict cell activation statue. We applied the method to classify lymphocyte videos of multiple groups. Experimental results demonstrate that the proposed method overcomes existing methods in accuracy and robustness.

Keywords: Deformable image registration, Image segmentation
Abstract: This paper discusses the development of a statistical shape model (SSM) of the skull from a South African population. A total of 16 skulls is used together with a reference from the Basel Face Model. A free-form deformation model is built using the reference model and the squared exponential kernel. The freeform model is used to establish dense correspondence across the skull sample and the in-correspondence skulls are then used to build an SSM. The validity of the SSM is assessed using the leave-one out cross-validation method with its generality ranging between 1.47 mm and 1.84 mm and, the specificity ranging between 1.74 mm and 2.11 mm.

Keywords: Cardiovascular and respiratory system modeling - Gas exchange models, Pulmonary and critical care - Bioengineering applications in Intensive care, Cardiovascular and respiratory system modeling - Cardiovascular-Respiratory Interactions
Abstract: Gaseous mixing in the anatomical deadspace with stimulation of respiratory ventilation through cardiogenic oscillations is an important physiological mechanism at the onset of apnea, which has been credited with various beneficial effects, e.g. reduction of hypercapnia during the use of low flow ventilation techniques. In this paper, a novel method is proposed to investigate the effect of these mechanisms in silico. An existing computational model of cardio-pulmonary physiology is extended to include the apneic state, gas mixing within the anatomical deadspace, insufflation into the trachea and cardiogenic oscillations. The new model is validated against data published in an experimental animal (dog) study that reported an increase in arterial partial pressure of carbon dioxide (PaCO2) during apnea. Computational simulations confirm that the model outputs accurately reproduce the available experimental data. This new model can be used to investigate the physiological mechanisms underlying clearance of carbon dioxide during apnea, and hence to develop more effective ventilation strategies for apneic patients.

Keywords: Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: Measuring diaphragmatic electromyography (EMGdi) provides an indirect quantification of neural respiratory drive and allows the delimitation of diaphragm neural activation and deactivation during inspiration. EMGdi recordings have been incorporated in novel modes of assisted mechanical ventilation, such as neurally adjusted ventilatory assist (NAVA), to trigger and cycle-off the ventilator. The EMGdi signal improves the assistance delivered by more conventional ventilatory modes, in which the ventilator is synchronized with the patient employing a pneumatic triggering. In this work, we evaluate the time delay between the onset and offset of inspiratory activity estimated from EMGdi and three respiratory mechanical signals: the respiratory flow (FL), the transdiaphragmatic pressure (Pdi) and the diaphragm length (Ldi) signals. To this purpose, these signals were acquired in three mongrel dogs surgically instrumented under general anesthesia. Onsets and offsets were estimated manually and by automatic algorithms on these signals. The highest delays were obtained between EMGdi and FL (100 ms) while the lowest delays were obtained between EMGdi and Pdi (8 ms). Moreover, differences between manual and automatic estimations showed a mean absolute error lower than 45 ms. In conclusion, our study points out that both EMGdi and Pdi signals detect the onset and offset of inspiratory activity earlier than the FL signal, and would therefore be better for the improvement of patient-ventilator synchrony.

Keywords: Pulmonary and critical care - Bioengineering applications in Intensive care, Pulmonary and critical care - Pulmonary function testing & instrumentation, Respiratory transport, mechanics and control - Respiratory variability
Abstract: The age dependence of the time-based capnogram from normal, healthy subjects has not been quantitatively characterized. The existence of age dependence would impact the development and operation of automated quantitative capnographic tools. Here, we quantitatively assess the relationship between normal capnogram shape and age. Capnograms were collected from healthy subjects, and physiologically-based features (exhalation duration, end-tidal CO2 and time spent at this value, normalized time spent at end-tidal CO2, end-exhalation slope, and instantaneous respiratory rate) were computationally extracted. The mean values of the individual features over 30 exhalations were linearly regressed against subject age, accounting for inter-feature correlation. After data collection, 154 of 178 subjects were eligible for analysis, with an age range of 3–78 years (mean age 39, std. dev. 20 years). The Bonferroni-corrected joint 95% confidence intervals (CIs) of the regression line slopes contained the origin for five of six features (the remaining CI was only slightly offset from the origin). The associated individual r2 values for the regressions were all below 0.07. We conclude that age is not a significant explanatory factor in describing variations in the shape of the normal capnogram. This finding could be exploited in the design of automated methods for quantitative capnogram analysis across a range of ages.

Keywords: Pulmonary and critical care - Bioengineering applications in Intensive care, Pulmonary and critical care - Ventilatory Assist Devices
Abstract: Electrical impedance tomography (EIT) allows functional imaging of regional lung ventilation for real-time bedside monitoring of mechanically ventilated patients. Images showing time-changes of regional air distributions in the lungs can provide valuable diagnostic information for lung protective mechanical ventilation. This paper reports in vivo porcine imaging experiments of regional lung ventilation using a 16-channel parallel EIT system. Real-time time-difference chest images of 10 animals were reconstructed during mechanical ventilations with a temporal resolution of 50 frame/s. Analyzing the images together with the airway volume-pressure information from the mechanical ventilator, we could successfully produce regional compliance images at PEEP (positive end expiratory pressure) titration. From in vivo animal experiments, we propose the method as a continuous monitoring means for LPV (lung protective ventilation).

Keywords: Pulmonary and critical care - Bioengineering applications in Intensive care, Pulmonary and critical care - Pulmonary disease
Abstract: Abstract—Adequate medical treatment of the Acute Respiratory Distress Syndrome is still challenging since patient-individual aspects have to be taken into account. Lung protective ventilation and hemodynamic stability have always been two of the most crucial aims of intensive care therapy. For both aspects, a continuous – preferably non-invasive – monitoring is desirable that is available at the bedside. Unfortunately, there is no technique clinically established yet, that provides both measurement of cardiac stroke volume and ventilation dynamics in real-time. Electrical Impedance Tomography (EIT) is a promising technique to close this gap. The aim of the study was to investigate if stroke volume can be estimated by a self-developed software using EIT-based image analysis. In addition, two EIT-derived parameters, namely Global Inhomogeneity Index (GII) and Impedance Ratio (IR), were calculated to evaluate homogeneity of air distribution. Experimental acute lung injury (ALI) was provoked in seven female pigs (German Landrace) by lipopolysaccharide (LPS). All animals suffered from experimental ALI 3 to 4 hours after LPS infusion. At defined time points, respiratory and hemodynamic parameters, blood gas analyses and EIT- recordings were performed. Eight hours after ALI, animals were euthanized. Stroke volume, derived from pulmonary artery catheter (PAC), decreased continuously up to four hours after ALI. Then, stroke volume increased slightly. Stroke volume, derived from the self-developed tool, showed the same characteristics (p=0.047, r = 0.365). In addition to the GII and IR individually, both classified scores showed a high correlation with the Horowitz Index, defined as paO2/FiO2. To conclude, EIT-derived measures enabled a reliable estimation of cardiac stroke volume and regional distribution of ventilation.

Keywords: Retinal vascular imaging, Retinal imaging, Image feature extraction
Abstract: Retinal vessel segmentation forms an essential element of automatic retinal disease screening systems. The development of multimodal imaging system with IR-SLO and OCT could help in studying the early stages of retinal disease. The advantages of IR-SLO to examine the alterations in the structure of retina and direct correlation with OCT can be useful for assessment of various diseases. This paper presents an automatic method for segmentation of IR-SLO fundus images based on the combination of morphological filters and image enhancement techniques. As a first step, the retinal vessels are contrasted using morphological filters followed by background exclusion using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Bilateral filtering. The final segmentation is obtained by using Isodata technique. Our approach was tested on a set of 26 IR-SLO images and results were compared to two set of gold standard images. The performance of the proposed method was evaluated in terms of sensitivity, specificity and accuracy. The system has an average accuracy of 0.90 for both the sets.

Keywords: Retinal vascular imaging, Retinal imaging
Abstract: This paper introduces the use of fluid-dynamic modeling to determine the connectivity of overlapping venous and arterial vessels in fundus images. Analysis of the retinal vascular network may provide information related to systemic and local disorders. However, the automated identification of the vascular trees in retinal images is a challenging task due to the low signal-to-noise ratio, nonuniform illumination and the fact that fundus photography is a projection on to the imaging plane of three-dimensional retinal tissue. A zero-dimensional model was created to estimate the hemodynamic status of candidate tree configurations. Simulated annealing was used to search for an optimal configuration. Experimental results indicate that simulated annealing was very efficient on test cases that range from small to medium size networks, while ineffective on large networks. Although for large networks the nonconvexity of the cost function and the large solution space made searching for the optimal solution difficult, the accuracy (average success rate = 98.35%), and simplicity of our novel approach demonstrate its potential effectiveness in segmenting retinal vascular trees.

Keywords: Retinal imaging, Image segmentation, Image feature extraction
Abstract: Retinal image analysis has been widely used for early detection and diagnosis of multiple systemic diseases. Accurate vessel extraction in retinal image is a crucial step towards a fully automated diagnosis system. This work affords an efficient unsupervised method for extracting blood vessels from retinal images using Gabor Wavelet (GW) with automatic thresholding. Green channel image is extracted from color retinal image and used to produce Gabor feature image using GW. Both green channel image and Gabor feature image undergo vessel-enhancement step in order to highlight blood vessels. Next, the two vessel-enhanced images are transformed to binary images using automatic thresholding before combined to produce the final vessel output. Combining the images results in significant improvement of blood vessel extraction performance compared to using individual image. Effectiveness of the proposed method was proven via comparative analysis with existing methods validated using publicly available database, DRIVE.

Keywords: Retinal vascular imaging, Image segmentation
Abstract: Retinal vessel segmentation takes a significant part in an automated diabetic retinopathy screening task. However, this can be a challenging job because of the low contrast retinal images and the presences of retinal pathologies. Hence, in this paper, we propose a novel matched filter based on the modified Chebyshev type I function for retinal blood vessels candidates detection. The proposed method is combined with the pre-processing and the post-processing phases to have a complete retinal vessel segmentation scheme. The retinal images from the DRIVE and STARE databases, which are equipped with the ground truths are used to evaluate our proposed method in the segmentation scheme. Using our method, the achieved average levels of sensitivity, specificity, and accuracy are 0.756, 0.973, and 0.954, for the DRIVE database, and 0.731, 0.972, and 0.953, for the STARE database, being better than other presented methods. The high results indicate that our method is reliable to be used in an automated detection tool for diabetic retinopathy.

Keywords: Retinal vascular imaging, Retinal imaging
Abstract: The pattern of blood vessels in the eye is unique to each person because it rarely changes over time. Therefore, it is well known that retinal blood vessels are useful for biometrics. This paper describes a biometrics method using the Jaccard similarity coefficient (JSC) based on blood vessel regions in retinal image pairs. The retinal image pairs were rough matched by the center of their optic discs. Moreover, the image pairs were aligned using the Iterative Closest Point algorithm based on detailed blood vessel skeletons. For registration, perspective transform was applied to the retinal images. Finally, the pairs were classified as either correct or incorrect using the JSC of the blood vessel region in the image pairs. The proposed method was applied to temporal retinal images, which were obtained in 2009 (695 images) and 2013 (87 images). The 87 images acquired in 2013 were all from persons already examined in 2009. The accuracy of the proposed method reached 100%.

Keywords: Retinal imaging, Deformable image registration
Abstract: This work regards an investigation of the accuracy of a state-of-the-art, keypoint-based retinal image registration approach, as to the type of keypoint features used to guide the registration process. The employed registration approach is a local method that incorporates the notion of a 3D retinal surface imaged from different viewpoints and has been shown, experimentally, to be more accurate than competing approaches. The correspondences obtained between SIFT, SURF, Harris-PIIFD and vessel bifurcations are studied, either individually or in combinations. The combination of SIFT features with vessel bifurcations was found to perform better than other combinations or any individual feature type, alone. The registration approach is also comparatively evaluated against representative methods of the state-of-the-art in retinal image registration, using a benchmark dataset that covers a broad range of cases regarding the overlap of the acquired images and the anatomical characteristics of the imaged retinas.

Keywords: Physiological systems modeling - Signal processing in physiological systems
Abstract: A surface electromyography (sEMG) signal typically results from the electrical activities of many muscle fibers, and can be utilized as a signal source in prostheses due to its abundance of movement information. This paper proposes an sEMG--detection circuit for the acquisition of the controlling signal in EMG-Bridge (EMGB) systems. The detection circuit mainly comprises a preamplifier, a driven right leg (DRL) circuit, a high-pass filter (HPF), a low-pass filter (LPF), and a gain adjustable amplifying circuit. The common-mode rejection ratio (CMRR) of the circuit is higher than 120 dB, the input impedance is greater than 100 MΩ, the passband range is 20~450 Hz, and the frequency attenuation in stopband is not less than 120dB/dec.

Keywords: Physiological systems modeling - Signals and systems, Physiological systems modeling - Signal processing in simulation, Physiological systems modeling - Signal processing in physiological systems
Abstract: This study investigates the effect of preparation warning on the functional connectivity between forearm muscles during movement’s initiation. The Normalized Mutual Information (NMI) is the used indicator. It is calculated for muscle pairs activated during a handgrip exercice. They are innervated by the same nerve or by different nerves. Results show that: i) NMI is more important when muscles pairs are stimulated by the same nerve during both preparation and motor activities. ii) NMI is higher during preparation comparing to motor activity. iii) The functional connectivity during contraction’s beginning increases when a preparation warning is given. We conclude that muscles put together their effort to initiate the movement especially when participants prepare in advance their activity.

Keywords: Signal pattern classification, Data mining and processing - Pattern recognition, Principal and independent component analysis - Blind source separation
Abstract: It is challenging to obtain good results for hand movements classification. Previous studies expended efforts on filters for sEMG data, feature extraction and classifier algorithms to achieve the best results. This paper proposes the insertion of a step in the classification process that selects which features to use in training aiming to increase accuracy and performance. Feature selection was previously used in other classification tasks but is new in wrist/fingers movements classification. Obtained results were positives as the performance gain is huge (39 to 53 features out of 144 are used for classification) and accuracy reach promising values (above 90% for some subjects).

Keywords: Physiological systems modeling - Signal processing in physiological systems, Time-frequency and time-scale analysis - Time-frequency analysis, Nonlinear dynamic analysis - Biomedical signals
Abstract: In recent years, active prosthetic legs have been developed and deployed commercially that help amputees to initiate gait with less effort and more symmetry in the pattern. However, the process of initial set up and tuning is highly time and cost consuming. It requires prosthetic experts to observe the gait and the feedback from amputees to manually tune the parameters subjectively. In this study, an electromyography (EMG)-based energy expenditure optimization method was presented to automatically tune the prosthetic limb. For this purpose, a wide variety of lower body muscles were observed and the energy expenditure was modeled based on their electrical activity. The tuning optimization was implemented based on a grid-searching protocol designed in this study. This method resulted in a power value comparable to manual tuning, which provided enough force to facilitate gait for amputees. This study shows the feasibility of automatic tuning and removal of the need for referral to an expert.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Signal pattern classification
Abstract: We studied the effects of muscle fatigue on the Autonomic Nervous System (ANS) dynamics. Specifically, we monitored the electrodermal activity (EDA) on 32 healthy subjects performing isometric biceps contraction. As assessed by means of an electromyography (EMG) analysis, 15 subjects showed muscle fatigue and 17 did not. EDA signals were analyzed using the recently proposed cvxEDA model in order to decompose them into their phasic and tonic components and extract effective features to study ANS dynamics. A statistical comparison between the two groups of subjects was performed. Results revealed that relevant phasic EDA features significantly increased in the fatigued group. Moreover, a pattern recognition system was applied to the EDA dataset in order to automatically discriminate between fatigued and non-fatigued subjects. The proposed leave-one-subject-out KNN classifier showed an accuracy of 75.69%. These results suggest the use of EDA as correlate of muscle fatigue, providing integrative information to the standard indices extracted from the EMG signals.

Keywords: Principal component analysis, Coupling and synchronization - Coherence in biomedical signal processing, Partial and total coherence
Abstract: Isolated isometric, concentric and eccentric forms of muscle action (MA) have been receiving the most attention. Nevertheless natural MA frequently involves the use of a preceding eccentric MA as a form of potentiation of immediate concentric MA, in what is designated as muscle stretch-shortening cycle (SSC). The most frequently applied protocols for the evaluation of SSC on vertical jumps are by virtue of their reproducibility and control of experimental conditions, squat jump (SJ) without countermovement (CM), countermovement jump (CMJ) with long CM and drop jump (DJ) with short CM. The methods used to extract information and relationship of the captured signals present a high diversity, with the question about the consistency of the methods and obtained results. The objective of this study is to evaluate the consistency of the analysis and results by applying different EMGs signal analysis techniques related to strategic muscle groups of the lower limbs at different CM evaluated in vertical jumps. Raw sEMG signals of 5 lower limb muscles of 6 subjects, captured at 1000 Hz during SJ, CMJ and DJ were rectified, filtered and obtained their envelope, and then correlated (CR) for detection of synergistic, agonist and antagonist activity, applied principal component analysis (PCA) for the detection of uncorrelated components explaining maximum variability and normalized cross-correlation (CCRN) for detection of maximum correlations and time lag. CR of EMG envelopes presented higher coactivities (CoA) in DJ relative to SJ and these CoA superior to CMJ with greater synergy in DJ relative to SJ and CMJ that present several loop cycles corresponding to time lag of activity. CCRN of the EMG envelopes presented also higher CoA in DJ when compared to SJ and both higher CoA to CMJ. PCA allowed to detect a principal component (PC) explaining 92.2% DJ EMG variability, 90.6% SJ and 78.7% CMJ, with 2nd PC explaining 4.9% DJ, 6.7% SJ and 15.3% CMJ variability.

Keywords: Signal pattern classification, Nonlinear dynamic analysis - Biomedical signals, Data mining and processing - Pattern recognition
Abstract: In the practical use of an electromyography (EMG) pattern-recognition based myoelectric prosthesis, the variation of force levels to do a motion would be inevitable, which will cause a change of EMG patterns. Therefore, the force variation will decay the performance of a trained classifier. In this study, the common spatial pattern (CSP) method was proposed with an attempt to improve the robustness of EMG-PR based classifier against force variation. The EMG signals were acquired from three able-bodied subjects when they were performing the motions at low, medium, and high force levels, respectively. And in the pattern recognition, CSP features were extracted from the EMG signals for motion classification. By comparing the classification accuracies between the CSP and the commonly used time-domain (TD) features, the CSP features showed a better robustness against force variation with an increment of 5.3% of the average classification accuracy. Especially, the classification accuracy of a classifier was 84.2% when tested at low force level by using CSP features, which was 18.5% higher than that of the TD features. These preliminary results suggest that using CSP features may increase the robustness of EMG-based myoelectric control.

Keywords: Signal pattern classification
Abstract: As motion artefacts are a major problem with electromyography sensors, a new algorithm is developed to differentiate artefacts to contraction EMG. The performance of myoelectric prosthesis is increased with this algorithm. The implementation is done for an ultra-low-power microcontroller with limited calculation resources and memory. The Short Time Fourier Transformation is used to enable real-time application. The sum of the differences (SOD) of the currently measured EMG to a reference contraction EMG is calculated. The SOD is a new parameter introduced for EMG classification. The satisfactory error rates are determined by measurements done with the capacitively coupling EMG prototype, recently developed by the research group.

Keywords: Signal pattern classification
Abstract: Measuring attention from electroencephalogram (EEG) has found applications in the treatment of Attention Deficit Hyperactivity Disorder (ADHD). It is of great interest to understand what features in EEG are most representative of attention. Intensive research has been done in the past and it has been proven that frequency band powers and their ratios are effective features in detecting attention. However, there are still unanswered questions, like, what features in EEG are most discriminative between attentive and non-attentive states? Are these features common among all subjects or are they subject-specific and must be optimized for each subject? Using Mutual Information (MI) to perform subject-specific feature selection on a large data set including 120 ADHD children, we found that besides theta beta ratio (TBR) which is commonly used in attention detection and neurofeedback, the relative beta power and theta/(alpha+beta) (TBAR) are also equally significant and informative for attention detection. Interestingly, we found that the relative theta power (which is also commonly used) may not have sufficient discriminative information itself (it is informative only for 3.26% of ADHD children). We have also demonstrated that although these features (relative beta power, TBR and TBAR) are the most important measures to detect attention on average, different subjects have different set of most discriminative features.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification
Abstract: In this paper, we present a new modular approach for detection of inter-ictal spikes in intracranial iEEG recordings from patients that are suffering from pharmaco-resistant form of epilepsy. This new approach is presented in the form of a detection framework consisting of three primary modules: first level detector, second level feature extractor, and third level detection classifier, where each module is responsible for a specific functionality. This detection framework can be perceived as a three slot system, where modules can be easily plugged in their slots and replaced by a different module or implementation on demand, in order to adapt the quality of detection (measured in terms of sensitivity, precision or inter-recording adaptability) and computational cost. Using complex real-world data sets it was confirmed that the proposed framework provides highly sensitive and precise detection, while it also significantly reduces the computation time.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Coupling and synchronization - Nonlinear coupling, Nonlinear dynamic analysis - Biomedical signals
Abstract: Dementia due to Alzheimer’s disease (AD) is a common disorder with a great impact on the patients’ quality of life. The aim of this pilot study is to characterize spontaneous electroencephalography (EEG) activity in dementia due to AD using bispectral analysis. Five minutes of EEG activity were recorded from 17 patients with moderate dementia due to AD and 19 age-matched controls. Bispectrum results revealed that AD patients are characterized by an increase of phase coupling at low frequencies in comparison with controls. Additionally, some bispectral features calculated from the bispectrum showed significant differences between both groups (p < 0.05, Mann-Whitney U test with Bonferroni’s correction). Finally, a stepwise logistic regression analysis with a leave-one-out cross-validation procedure was used for classification purposes. An accuracy of 86.11% (sensitivity = 88.24%; specificity =84.21%) was achieved. This study suggests the usefulness of bispectral analysis to provide further insights into the underlying brain dynamics associated with AD.

Keywords: Brain functional imaging - NIR, Brain functional imaging - Connectivity and information flow
Abstract: Physiological fluctuations are commonly present in functional studies of hemodynamic response such as functional magnetic resonance imaging (fMRI) and functional near-infrared spectroscopy (fNIRS). However, the effects of these signals in neural mechanisms are not fully understood. Thus, the aim of this study is to propose that frequency-specific networks exist in the somatosensory region within the frequency range of physiological fluctuations. We used a wavelet coherence approach to identify functional connectivity between cortical regions. Based on the spectral response, four frequency bands were identified: cardiac (0.8-1.5 Hz), respiration (0.16-0.6 Hz), low frequency oscillations (LFO) (0.04-0.15 Hz), and very low frequency oscillations (VLFO) (0.01-0.04 Hz). Eight cortical networks were revealed after ipsilateral and contralateral analysis to evaluate connectivity in each frequency band. The ANOVA analysis proved the adequacy of the connectivity map for all frequencies bands. Finally, these findings suggest possible frequency-specific organizations within the frequency bands of physiological fluctuations in the resting human brain.

Keywords: Signal pattern classification, Physiological systems modeling - Signal processing in physiological systems
Abstract: Identification of brain states measured with electrophysiological methods such as electroencephalography and local field potential (LFP) recordings is of great importance in numerous neuroscientific applications. For instance, in Brain Computer Interface, in the diagnosis of neurological disorders as well as to investigate how brain rhythms stem from synchronized physiological mechanisms (e.g., memory and learning). In this work, we propose a fully automated method with the aim of partitioning LFP signals into stationary segments as well as classifying each detected segment into three different classes (delta, regular theta or irregular theta rhythms). Our approach is computationally efficient since the process of detection and partition of signals into stationary segments is only based on two features (the variance and the so-called spectral error measure) and allow the classification at the same time. We developed the algorithm upon analyzing six anesthetized rats, resulting in a true positive rate of 97.5%, 91.8% and 79.1% in detecting delta, irregular theta and regular theta rhythms, respectively. This preliminary quantitative evaluation offers encouraging results for further research.

Keywords: Signal pattern classification, Physiological systems modeling - Signal processing in physiological systems
Abstract: We discuss the adaptation of preexisting Convolution Kernel Compensation (CKC) surface electromyogram (EMG) decomposition technique to dynamic muscle contractions. In particular, three different algorithms for segmentation of motor unit (MU) spike trains into MU firings are discussed and mutually compared on synthetic dynamic surface EMG. The first segmentation algorithm employs a priori knowledge of the regularity of MU firings. The second one builds on K-means classification of MU spikes [6], whereas the third one combines both the regularity of MU firings and the previously introduced Pulse-to-Noise Ratio (PNR). On average, 5.5 ± 0.6 MUs were identified with sensitivity of 88.4 % ± 17.0 %, 83.8 % ± 16.7 % and 90.7 % ± 15.1 % for the first, the second and the third segmentation algorithm, respectively, demonstrating the feasibility of MU identification in moderate dynamic muscle contractions. In our tests, the third segmentation approach demonstrated superior accuracy in MU identification.

Keywords: Independent component analysis, Signal pattern classification, Nonlinear dynamic analysis - Biomedical signals
Abstract: Coronary Artery Disease (CAD) is the most leading cardiovascular disease (CVD), which results due to buildup of plaque inside the coronary arteries. The CAD and Normal Sinus Rhythm (NSR) heartbeats can be discriminated and diagnosed noninvasively using the standard tool Electrocardiogram (ECG). However, manual diagnosis of ECG is tiresome and time consuming task, due to complex nature and unseen nonlinearities of ECG. Hence an automated system plays a substantial role. In this study, CAD and NSR heartbeats are discriminated and diagnosed using Higher-Order Statistics (HOS) cumulants features. Data reduction is achieved using Principal Components Analysis (PCA) and then medically significant features (p-value<0.05) are subjected for classification using Random Forest (RAF) and Rotation Forest (ROF) ensemble classification method. Developed system is efficient helps in mass screening and telemonitoring applications.

Keywords: Data mining and processing - Pattern recognition, Data mining and processing in biosignals
Abstract: Cardiac arrhythmia or irregular heartbeats are an important feature to assess the risk on sudden cardiac death and other cardiac disorders. Automatic classification of irregular heartbeats is therefore an important part of ECG analysis. We propose a tensor-based method for single- and multi-channel irregular heartbeat classification. The method tensorizes the ECG data matrix by segmenting each signal beat-by-beat and then stacking the result into a third-order tensor with dimensions channel x time x heartbeat. We use the multilinear singular value decomposition to model the obtained tensor. Next, we formulate the classification task as the computation of a Kronecker Product Equation. We apply our method on the INCART dataset, illustrating promising results.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Neural networks and support vector machines in biosignal processing and classification, Signal pattern classification
Abstract: Rett syndrome (RTT) is a severe neurodevelopmental disorder that can cause pervasive wakeful respiratory disturbances that include tachypnea, breath-holding, and central apnea. Quantitative analysis of these respiratory disturbances in RTT is considered a promising outcome measure for clinical trials. Currently, machine learning methodologies have not been employed to automate the classification of RTT respiratory disturbances. In this paper, we propose using temporal, flow, and autocorrelation features taken from the respiratory inductance plethsymography chest signal. We tested the performance of six classifiers including: Support Vector Machine, Restricted-Boltzmann-Machine, Back-propagation, Levenberg-Marquardt, and Decision-Fusion. We evaluate this classification in two modalities: (1) a subject-independent modality (leave-one-subject-out) obtaining the best F1 score in 93.67%, and (2) a trial-independent modality (leave-one-trial-out per subject) obtaining the best F1 score in 78.21% respectively.

Keywords: Signal pattern classification
Abstract: Human authentication based on electrocardiogram (ECG) has been a remarkable issue for recent ten years. This paper proposed an authentication technology with the ECG data recorded after the harsh exercise. 55 subjects voluntarily attended to this experiment. A stepper was used as an exercise equipment. The subjects are asked to do stepper for 5 minutes and their ECG signals are acquired before and after the exercise in rest, sitting posture. Linear discriminant analysis (LDA) was used for both feature extraction and classification. Even though, within the first 1 minute recording, the subject recognition accuracy was 59.64%, which is too low to utilize, after one minute the accuracy was higher than 90% and it increased up to 96.22% within 5 minutes, which is plausible to use in authentication circumstances. Therefore, we have concluded that ECG authentication techniques will be able to be used after 1 minute of catching breath.

Keywords: Data mining and processing - Pattern recognition, Signal pattern classification
Abstract: Emotions modulate ECG signals such that they might affect ECG-based biometric identification in real life application. It motivated in finding good feature extraction methods where the emotional state of the subjects has minimum impacts. This paper evaluates feature extraction based on bivariate empirical mode decomposition (BEMD) for biometric identification when emotion is considered. Using the ECG signal from the Mahnob-HCI database for affect recognition, the features were statistical distributions of dominant frequency after applying BEMD analysis to ECG signals. The achieved accuracy was 99.5% with high consistency using kNN classifier in 10-fold cross validation to identify 26 subjects when the emotional states of the subjects were ignored. When the emotional states of the subject were considered, the proposed method also delivered high accuracy, around 99.4%. We concluded that the proposed method offers emotion-independent features for ECG-based biometric identification. The proposed method needs more evaluation related to testing with other classifier and variation in ECG signals, e.g. normal ECG vs. ECG with arrhythmias, ECG from various ages, and ECG from other affective databases.

Keywords: Signal pattern classification
Abstract: Biometrics such as ECG provides a convenient and powerful security tool to verify or identify an individual. However, one important drawback of biometrics is that it is irrevocable. In other words, biometrics cannot be re-used practically once it is compromised. Cancelable biometrics has been investigated to overcome this drawback. In this paper, we propose a cancelable ECG biometrics by deriving a generalized likelihood ratio test (GLRT) detector from a composite hypothesis testing in randomly projected domain. Since it is common to observe performance degradation for cancelable biometrics, we also propose a guided filtering (GF) with irreversible guide signal that is a non-invertibly transformed signal of ECG authentication template. We evaluated our proposed method using ECG-ID database with 89 subjects. Conventional Euclidean detector with original ECG template yielded 93.9% PD1 (detection probability at 1% FAR) while Euclidean detector with 10% compressed ECG (1/10 of the original data size) yielded 90.8% PD1. Our proposed GLRT detector with 10% compressed ECG yielded 91.4%, which is better than Euclidean with the same compressed ECG. GF with our proposed irreversible ECG template further improved the performance of our GLRT with 10% compressed ECG up to 94.3%, which is higher than Euclidean detector with original ECG. Lastly, we showed that our proposed cancelable ECG biometrics practically met cancelable biometrics criteria such as efficiency, re-usability, diversity and non-invertibility.

Keywords: Data mining and processing in biosignals, Signal pattern classification, Physiological systems modeling - Multivariate signal processing
Abstract: This paper introduces a new unsupervised method for the clustering of physiological data into health states based on their similarity. We propose an iterative hierarchical clustering approach that combines health states according to a similarity constraint to new arbitrary health states. We applied our method to experimental data in which the physical strain of subjects was systematically varied. We derived health states based on parameters extracted from ECG data. The occurrence of health states shows a high temporal correlation to the experimental phases of the physical exercise. We compared our method to other clustering algorithms and found a significantly higher accuracy with respect to the identification of health states.

Keywords: Time-frequency and time-scale analysis - Wavelets, Signal pattern classification, Data mining and processing - Pattern recognition
Abstract: Recognizing emotion from EEG signals is a complicated task that requires complex features and a substantial number of EEG channels. Simple algorithms to analyse the feature and reduce the EEG channel number will give an indispensable advantages. Therefore, this study explores a combination of wavelet entropy and average wavelet coefficient (WEAVE) as a potential EEG-emotion feature to classify valence and arousal emotions with the advantage of the ability to identify the occurrence of a pattern while at the same time identify the shape of a pattern in EEG emotion signal. The complexity of the feature was reduced using the Normalized Mutual Information (NMI) method to obtain a reduced number of channels. Classification with the WEAVE feature achieved 76.8% accuracy for valence and 74.3% for arousal emotion, respectively. The analysis with NMI shows that the WEAVE feature has linear characteristics and offers possibilities to reduce the EEG channels to a certain number. Further analysis also reveals that detection of valence emotion with reduced EEG channels has a different combination of EEG channels compared to arousal emotion.

Keywords: Time-frequency and time-scale analysis - Time-frequency analysis, Signal pattern classification
Abstract: In the current study, we tested a proposed method for fast spike detection using a general-purpose computer. First, we performed eigenvalue analysis using a gradient calculated from two neighboring samples to detect high-amplitude negative peaks. Clustering was performed to classify detected peaks by considering amplitude distribution at scalp electrodes. Negative peaks were scored by considering electrodes in the detection process and the cluster to which each peak belonged. Spikes were detected using two parameters: score threshold, and the number of clusters. We then used precision and recall to eliminate overestimation of the performance of the method. The results revealed a trade-off between precision and recall. Recall showed a maximum average value of 0.90 in two subjects. In contrast, average precision was 0.21, and the false positive rate was almost four times higher than the true positive rate. Analysis of required processing time revealed that our method could complete spike detection in approximately one-eighth of the recording time, using a general-purpose computer.

Keywords: Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification
Abstract: For deep learning on image data, a common approach is to augment the training data by artificial new images, using techniques like moving windows, scaling, affine distortions, and elastic deformations. In contrast to image data, electroencephalographic (EEG) data suffers even more from the lack of sufficient training data. Methods: We suggest and evaluate rotational distortions similar to affine/rotational distortions of images to generate augmented data. Results: Our approach increases the performance of signal processing chains for EEG-based brain-computer interfaces when rotating only around y- and z-axis with an angle around ±18 degrees to generate new data. Conclusion: This shows that our processing efficient approach generates meaningful data and encourages to look for further new methods for EEG data augmentation.

Keywords: Time-frequency and time-scale analysis - Wavelets, Signal pattern classification, Physiological systems modeling - Signal processing in physiological systems
Abstract: Epilepsy is a neurological disorder for which the electroencephalogram (EEG) is the most important diagnostic tool. In particular, this diagnosis heavily depends on the detection of interictal (between seizures) paroxysmal epileptic discharges (IPED) in the EEG. This is a time-consuming task requiring significant training and experience. Automatic detection of these EEG patterns would greatly assist visual inspections of human readers. We present a new method, which allows automatic detection of IPED based on discrete wavelet decomposition and a random forest classifier. The algorithm was trained and cross validated using 17 subjects with scalp EEG and 10 subjects with intracranial EEG. The performance of this method reached 62% recall and 26% precision for surface EEG subjects and 63% recall and 53% precision for intracranial EEG subjects. Thus, the method hereby proposed has great potential for diagnosis support in clinical environments.

Keywords: Time-frequency and time-scale analysis - Empirical mode decomposition in biosignal analysis, Nonlinear dynamic analysis - Biomedical signals
Abstract: Through the investigation of the difference of approximate entropy (ApEn) and Hilbert-Huang Transform Marginal spectrum entropy (HHTMSEn) of the Electroenceph- Alogram (EEG) signals of specific low back pain (SLBP) patients before and after the massage, we wanted to reveal the impact of the massage in the brain, and provided the basis for the treatment of SLBP patients. We recruited twenty-six SLBP patients and collected their spontaneous EEG signals before and after the massage. Firstly, we analyzed the ApEn and HHTMSEn of 14 channels before and after the massage, and results showed that values of ApEn and HHTMSEn after the massage were less than the values before the massage significantly. And then, we extracted δ, θ, α and β rhythms of the EEG signals, and analyzed the ApEn and HHTMSEn of the four rhythms before and after the massage. The results showed that the ApEn values of δ and α rhythms after the massage were significantly less than the values before the massage, and the HHTMSEn values of β rhythms were significantly less than the values before the massage, and this showed that the complexity of EEG signal was reduced with the relief of the pain after the massage therapy.

Keywords: Parametric filtering and estimation, Signal pattern classification
Abstract: To decode auditory attention from single-trial EEG recordings in an acoustic scenario with two competing speakers, a least-squares method has been recently proposed. This method however requires the clean speech signals of both the attended and the unattended speaker to be available as reference signals. Since in practice only the binaural signals consisting of a reverberant mixture of both speakers and background noise are available, in this paper we explore the potential of using these (unprocessed) signals as reference signals for decoding auditory attention in different acoustic conditions (anechoic, reverberant, noisy, and reverberant-noisy). In addition, we investigate whether it is possible to use these signals instead of the clean attended speech signal for filter training. The experimental results show that using the unprocessed binaural signals for filter training and for decoding auditory attention is feasible with a relatively large decoding performance, although for most acoustic conditions the decoding performance is significantly lower than when using the clean speech signals.

Keywords: Signal pattern classification - Genetic algorithms, Time-frequency and time-scale analysis - Wavelets, Physiological systems modeling - Signal processing in physiological systems
Abstract: Emotion classification is one of the state-of-the-art topics in biomedical signal research, and yet a significant portion remains unknown. This paper offers a novel approach with a combined classifier to recognise human emotion states based on electroencephalogram (EEG) signal. The objective is to achieve high accuracy using the combined classifier designed, which categorises the extracted features calculated from time domain features and Discrete Wavelet Transform (DWT). Two innovative designs are involved in this project: a novel variable is established as a new feature and a combined SVM and HMM classifier is developed. The result shows that the joined features raise the accuracy by 5% on valence axis and 1.5% on arousal axis. The combined classifier can improve the accuracy by 3% comparing with SVM classifier. One of the important applications for high accuracy emotion classification system is offering a powerful tool for psychologists to diagnose emotion related mental diseases and the system developed in this project has the potential to serve such purpose.