Keywords: Brain functional imaging - EEG, Neural signal processing, Human performance - Sleep
Abstract: Newborn babies, particularly preterms, can exhibit early deviations in sleep maturation as seen by Electroencephalogram (EEG) recordings. This may be indicative of cognitive problems by school-age. The current ‘clinically driven’ approach uses separate algorithms to first extract sleep states and then predict EEG ‘brain-age’. Maturational deviations are identified when the brain-age no longer matches the Postmenstrual Age (PMA, the age since the last menstrual cycle of the mother). However, the PMA range where existing sleep staging algorithms perform optimally, is limited, which subsequently limits the PMA range for brain-age prediction.

We introduce a Bayesian Parametric Model (BPM) as a single end-to-end solution to directly estimate brain-age, modelling for sleep state maturation without requiring a separately optimized sleep staging algorithm. Comparison of this model with a traditional multi-stage approach, yields a similar Krippendorff’s alpha = 0.92 (a performance measure ranging from 0 (chance agreement) to 1 (perfect agreement)) with the BPM performing better at younger ages <30 weeks PMA. The BPM’s potential to detect maturational deviations is also explored on a few preterm babies who were abnormal at 9 months follow-up.

Keywords: Brain functional imaging - MEG, Brain functional imaging - Mapping, Brain physiology and modeling - Cognition, memory, perception
Abstract: It is known that alpha-band rhythm during memory maintenance is enhanced by increasing memory load. This enhancement is generally thought to be caused by active inhibition of task-irrelevant visual inputs. During sequential memory processing, we previously found that alpha-band activity increases from beginning to midterm during memory encoding, and conversely decreases from midterm to ending. In the present study, we conducted two experiments to determine the spatial and functional role of alpha-band rhythm during sequential memory processing. The first experiment showed that alpha-band rhythm increased in the occipital brain region, suggesting that active inhibition of task-irrelevant visual inputs continues from midterm to ending of memory encoding. The second experiment, in which subjects could not anticipate the ending of the sequential presentation of memory items, demonstrated that alpha-band rhythm is suppressed in correspondence with preparation for memory recall. These results indicate that alpha-band rhythm is simultaneously modulated by multiple brain processes in sequential memory encoding.

Keywords: Brain functional imaging - Classification, Brain functional imaging - MEG, Neural signal processing
Abstract: Target detection during serial visual presentation tasks is an active research topic in the brain-computer interface (BCI) community as this type of paradigm allows to take advantage of event-related potentials (ERPs) through electroencephalography (EEG) recordings to enhance the accuracy of target detection. The detection of brain evoked responses at the single-trial level remains a challenging task and can be exploited in various applications. Typical non-invasive BCIs based on event-related brain responses use EEG. In clinical settings, brain signals recorded with magnetoencephalography (MEG) can be advantageously used thanks to their high spatial and temporal resolution. In this study, we address the problem of the relationships between behavioral performance and single-trial detection by considering a task with different levels of difficulty. We consider images of faces with six different facial expressions (anger, disgust, fear, neutrality, sadness, and happiness). We consider MEG signals recorded on ten healthy participants in six sessions where targets were one of the six types of facial expressions in each session. The results support the conclusion that a high performance can be obtained at the single-trial level (AUC=0.903+/-0.045), and that the performance is correlated with the behavioral performance (reaction time and hit rate).

Keywords: Brain functional imaging - EEG, Neurological disorders - Diagnostic and evaluation techniques, Neural signal processing
Abstract: Mild cognitive impairment (MCI) and Alzheimer's Disease (AD) affect millions worldwide, yet no curative treatments for these near-degenerative disorders have been developed to date. The current study aims to propose a non-invasive, cost-effective, early diagnostic protocol for individuals suffering with MCI in an outpatient setting. Elderly participants (n=11) were screened for MCI utilizing the Montreal Cognitive Assessment (MoCA) questionnaire preceding a visual stimuli task. Participants were presented with facial stimuli to elicit event-related potentials (ERP) while their cortical activity was recorded utilizing electroencephalogram (EEG). Combining regional neurophysiological biomarkers into a multi-dimensional feature space allowed for differentiation between healthy and MCI participants based on their respective MoCA scores. This study illustrates the feasibility of recording reliable EEG in an outpatient setting while presenting a novel method for diagnosing MCI in elderly (age>60) populations.

Keywords: Brain functional imaging - Connectivity and information flow, Neurological disorders - Psychiatric disorders, Brain functional imaging - NIR
Abstract: This paper reports a functional connectivity analysis at prefrontal cortex (PFC) during semantic verbal fluency task (SVFT) for three groups of elderly people, i.e. normal aging (NA), mild cognitive impairment (MCI) and mild Alzheimer’s disease (AD). Functional Near Infrared Spectroscopy (fNIRS) was used to measure neuronal activities. A new software algorithm was developed to process fNIRS signals and to derive the parameters of functional connectivity. The synchronization of oxygenated hemoglobin signals from paired channels was evaluated using their temporal correlation. Results from 61 subjects of experiment show that a general decline in functional connectivity from NA (edge count = 307) to AD (edge count = 170), and the laterality between left and right PFC became insignificant (p>0.01) at AD stage. Moreover, the NA group demonstrated a significantly higher clustering coefficient than the AD group (p<0.01), indicating the NA has higher regularity in brain network. Using semantic verbal fluency task, this work demonstrated fNIRS as a feasible measuring instrument to differentiate AD from NA based on functional connectivity, with clustering coefficient and laterality as suitable biomarkers.

Keywords: Brain functional imaging - Spatial-temporal dynamics, Sensory neuroprostheses - Somatosensory, Neural stimulation
Abstract: Brain plasticity after amputation is related to the short-term unmasking of latent synapses as well as the long-term reorganization due to the sprouting new synaptic connections. The cortical functional reorganization along the intact sensory pathway has been shown evoked by unilateral deafferentation. The cerebral blood flow (CBF) change serves as an important biomarker of the functional reorganization of brain. Using laser speckle contrast imaging (LSCI) technology, we performed a longitudinal study to unveil the cortical reorganization after forelimb amputation in rodent model, particularly along the intact pathway. Our results showed that the CBF response under electrical stimulation to the intact forepaw increased significantly at 9 hours after amputation in acute stage. While in chronic stage (> 14 days), the CBF response showed a similar pattern to control group. The results showed the dynamic brain functional response along the intact sensory pathway at different stages after amputation and indicated that cortical functional reorganization occurred within acute stage. Our work provided additional insights in understanding the inter-hemispheric functional changes from acute to chronic stages of amputation.

Keywords: Parametric filtering and estimation, Adaptive filtering, Nonlinear dynamic analysis - Nonlinear filtering
Abstract: In a complex auditory scene comprising multiple sound sources, humans are able to target and track a single speaker. Recent studies have provided promising algorithms to decode the attentional state of a listener in a competing-speaker environment from non-invasive brain recordings such as electroencephalography (EEG). These algorithms require substantial training datasets and often exhibit poor performance at temporal resolutions suitable for real-time implementation, which hinders their utilization in emerging applications such as smart hearing aids. In this work, we propose a real-time attention decoding framework by integrating techniques from Bayesian filtering, l1-regularization, state-space modeling, and Expectation Maximization, which is capable of producing robust and statistically interpretable measures of auditory attention at high temporal resolution. Application of our proposed algorithm to synthetic and real EEG data yields a performance close to the state-of-the-art offline methods, while operating in near real-time with a minimal amount of training data.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Physiological systems modeling - Signal processing in simulation, Parametric filtering and estimation
Abstract: Minimally-invasive continuous glucose monitoring (CGM) sensors have revolutionized perspectives in the treatment of type 1 diabetes (T1D). Their accuracy relies on an internal calibration function that transforms the raw, physically measured, electrical data into blood glucose concentration values. Usually, a unique, pre-determined, calibration functional is adopted, with parameters periodically updated in individual patients by using "gold standard" references suitably collected by finger prick devices. However, retrospective analysis of CGM data suggests that variability of sensor-subject characteristics is often inefficiently coped with. In the present study, we propose a conceptual Bayesian model-selection framework aimed at guaranteeing wide margins of flexibility for both the determination of the most appropriate calibration functional and the numerical values of its unknown parameters. The calibration model is determined among a finite specified set of candidates, each one depending on a set of unknown model parameters, for which a priori statistical expectations are available. Model selection is based on predictive distributions carrying out asymptotic calculations through Monte Carlo integration methods. Performance of the proposed approach is assessed on synthetic data generated by a well-established T1D simulation model.

Keywords: Time-frequency and time-scale analysis - Nonstationary processing, Physiological systems modeling - Multivariate signal processing, Kalman filtering
Abstract: A recent work (Kim et al. 2018) has reported a novel statistical modeling framework, the State-Space Multitaper (SSMT) method, to estimate time-varying spectral representation of non-stationary time series data. It combines the strengths of the multitaper spectral (MT) analysis paradigm with that of state-space {(SS)} models. In this current work, we explore a variant of the original SSMT framework {by imposing a smoothness promoting {SS} model to generate smoother estimates of power spectral densities for non-stationary data. Specifically, we assume that the continuous processes giving rise to {observations} in the {frequencies} of interest follow multiple independent Integrated Wiener Processes (IWP).} We use both synthetic data and electroencephalography (EEG) data collected from a human subject under anesthesia to compare the IWP-SSMT with the SSMT method and demonstrate the former's utility in yielding smoother descriptions of underlying processes. The original SSMT and IWP-SSMT can co-exist as a part of a model selection toolkit for nonstationary time series data.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Nonlinear dynamic analysis - Biomedical signals
Abstract: Electrodermal activity (EDA) is a measure of sympathetic activity using skin conductance that has applications in research and in clinical medicine. However, current EDA analysis does not have physiologically-based statistical models that use stochastic structure to provide nuanced insight into autonomic dynamics. Therefore, in this study, we analyzed the data of two healthy volunteers under controlled propofol sedation. We identified a novel statistical model for EDA and used a point process framework to track instantaneous dynamics. Our results demonstrate for the first time that point process models rooted in physiology and built upon inherent statistical structure of EDA pulses have the potential to accurately track instantaneous dynamics in sympathetic tone.

Keywords: Kalman filtering, Data mining and processing - Pattern recognition
Abstract: Predicting the severity and onset of depressive symptoms is of great importance. User-specific models have better performance than a general model but require significant amounts of training data from each individual, which is often impractical to obtain. Even when this is possible, there is a significant lag between the beginning of the data-collection phase and when the system is completely trained and thus able to start making useful predictions. In this study, we propose a transfer learning Bayesian modelling method based on a Markov Chain Monte Carlo (MCMC) sampler and Bayesian model averaging for dealing with the challenge of building user-specific predictive models able to make predictions of self-reported well-being scores with limited sparse training data. The evaluation of our method using real-world data collected within the NEVERMIND project showed a better predictive performance for the transfer learning model compared to conventional learning with no transfer.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Physiological systems modeling - Signal processing in simulation
Abstract: Transition dynamics between two states can help elucidate the behavior of sequential events in physiological signals. By detecting transitions between healthy and pathological states within individual patients, we can help clinicians focus attention on critical transitions, to either preemptively treat adverse events or to detect changes resulting from treatments. We introduce a novel application of single-point Bayesian online changepoint detection to predict clinical state transitions, and apply this framework to detecting pathological transitions in preterm infants with episodes of apnea and bradycardia. Bayesian analysis of sequential physiological events provides insights on how to objectively classify clinically important state transitions that can be triggered by external or intrinsic mechanisms.

Keywords: Image registration, segmentation, compression and visualization - Machine learning / Deep learning approaches, Image segmentation, Image analysis and classification - Digital Pathology
Abstract: In the past decades, the number of in vitro fertilization (IVF) procedures for the conception of a child has been rising continuously, however, the success rate of artificial insemination remained low. According to current statistics, large portion of unsuccessful IVF relates to some women’ factors. As the directly related female organ, the proper investigation of the uterus has primary importance. Namely, visible markers may indicate inflammations or other negative effects that jeopardize successful implantation. The purpose of this study is to support the observability of the uterus from this aspect by providing computer-aided tools for the extraction of its wall from video hysteroscopy. As for methodology, fully convolutional neural networks (FCNNs) are used for the automatic segmentation of the video frames to determine the region of interest. We provide the necessary steps for the applicability of the general deep learning framework for this specific task. Moreover, we increase segmentation accuracy with applying ensemble-based approaches at two levels. First, the predictions of a given FCNN are aggregated for the overlapping regions of subimages, which are derived from the splitting of the original images. Next, the segmentation results of different FCNNs are fused via a weighted combination model; optimization for adjusting the weights are also provided. Based on our experimental results, we have achieved 91.56% segmentation accuracy regarding the recognition of the uterus wall.

Keywords: Image segmentation
Abstract: Automated segmentation of the spleen in CT volumes is difficult due to variations in size, shape, and position of the spleen within the abdominal cavity as well as similarity of intensity values among organs in the abdominal cavity. In this paper we present a method for automated localization and segmentation of the spleen within axial abdominal CT volumes using trained classification models, active contours, anatomical information, and adaptive features. The results show an average Dice score of 0.873 on patients experiencing various chest, abdominal, and pelvic traumas taken at different contrast phases.

Keywords: Magnetic resonance imaging - Diffusion tensor, diffusion weighted and diffusion spectrum imaging, Brain imaging and image analysis, Image analysis and classification - Machine learning / Deep learning approaches
Abstract: Diffusion tensor imaging (DTI) has recently been added to the large scale of studies for Alzheimer’s Disease (AD) to investigate the White Matter (WM) defects that are not detectable using structural MRI. In this paper, we extracted Speeded Up Robust Features (SURF) and Scale Invariant Feature Transform (SIFT) features, based on the visual diffusion patterns of Fractional Anisotropy (FA), and Mean Diffusivity (MD) maps, to build bag-of-words AD-signature for the hippocampal area. The experiments were accomplished with a subset of participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset formed of AD patients (n=35), Early Mild Cognitive Impairment (EMCI) (n=6), Late Mild Cognitive Impairment (LMCI) (n=24) and cognitively healthy elderly Normal Controls (NC) (n=31). The preliminary studied experiments give promising results that would consider the proposed system as an accurate and useful tool to capture the AD leanness with accuracy of 87% and 89% for FA and MD maps respectively.

Keywords: Multivariate image analysis, Image classification, Brain imaging and image analysis
Abstract: Parkinson’s Disease (PD) is the most common motor neurodegenerative disease in elderly population. Transcranial sonography (TCS) has become a popular imaging tool for diagnosis of PD in clinical practice. Moreover, several pioneering work have developed the computer-aided diagnosis (CAD) for PD with the transcranial B-mode sonography (TBS). It is worth noting that TCS not only has the TBS modality, but also can image the blood flow of major cerebral arteries, which is named transcranial Doppler sonography (TDS). TDS also has been applied to evaluate PD patients with orthostatic hypotension. However, the TDS-based CAD for PD has not been investigated. Since TBS and TDS provide the complementary structural and functional information about brain, it is feasible to develop a multi-modal TCS-based CAD for PD by combining both TBS and TDS. Therefore, in this work, we propose a multiple kernel learning (MKL) based CAD for PD with multi-modal TCS imaging. Particularly, the statistical and texture features are extracted from the midbrain region from TBS images, and the features about blood flow are calculated from the spectrum curves in TDS. The multi-modal features are then fed to a MKL classifier for classification of PD. The experimental results show that the multi-modal TCS-based method outperforms both the single-modal TBS- and TDS-based algorithm, which suggests the feasibility and effectiveness of combining TBS and TDS for diagnosis of PD.

Keywords: Image registration, segmentation, compression and visualization - Machine learning / Deep learning approaches
Abstract: Colorectal cancer is one of the common cancers in the United States. Polyps are one of the major causes of colonic cancer, and early detection of polyps will increase the chance of cancer treatments. In this paper, we propose a novel classification of informative frames based on a convolutional neural network with binarized weights. The proposed CNN is trained with colonoscopy frames along with the labels of the frames as input data. We also used binarized weights and kernels to reduce the size of CNN and make it suitable for implementation in medical hardware. We evaluate our proposed method using Asu Mayo Test clinic database, which contains colonoscopy videos of different patients. Our proposed method reaches a dice score of 71.20% and accuracy of more than 90% using the mentioned dataset.

Keywords: Image registration, segmentation, compression and visualization - Volume rendering, Optical imaging and microscopy - Microscopy
Abstract: Colorectal cancer is one of the highest causes of cancer-related death, especially in men. Polyps are one of the main causes of colorectal cancer, and early diagnosis of polyps by colonoscopy could result in successful treatment. Diagnosis of polyps in colonoscopy videos is a challenging task due to variations in the size and shape of polyps. In this paper, we proposed a polyp segmentation method based on the convolutional neural network. Two strategies enhance the performance of the method. First, we perform a novel image patch selection method in the training phase of the network. Second, in the test phase, we perform effective post-processing on the probability map that is produced by the network. Evaluation of the proposed method using the CVC-ColonDB database shows that our proposed method achieves more accurate results in comparison with previous colonoscopy video-segmentation methods.

Keywords: Sleep - Obstructive sleep apnea
Abstract: Obstructive Sleep Apnoea (OSA) is common, under-diagnosed, and difficult to treat. It is a heterogeneous disorder with different risk factors, clinical presentations, pathophysiology and morbidity. Prediction plays a key role in OSA detection and treatment but current in-laboratory style assessment limits the usefulness of the role. Alternative low-cost, at-home screening methods are useful to circumvent the need for diagnosis by overnight sleep studies. Precision medicine principles based on the framework of P4 medicine (prediction, prevention, personalised and participation) are the key to good management in OSA as they permit an understanding of disease phenotypes.

Keywords: Sleep - Obstructive sleep apnea, Sleep - Periodic breathing & central apnea
Abstract: We suggest using electrical impedance tomography (EIT) imaging of the chest and upper airway as a new method of apnea and hypopnea diagnosis. The method can be implemented as a home sleep test (HST) device or a supplement to polysomnography (PSG). We describe the development of a portable HST device including real-time EIT imaging of tidal volume changes, nasal pressure, respiratory effort, SpO2, ECG, and body position. The preliminary results of comparing its performance as a HST device are presented from a comparative studies with a PSG device. A new EIT device for real-time upper airway imaging during natural sleep will be described to provide new information about the obstruction site. Future studies of clinical trials will be proposed.

Keywords: Cardiovascular, respiratory, and sleep devices - Smart systems, Cardiovascular, respiratory, and sleep devices - Monitors, Sleep - Obstructive sleep apnea
Abstract: In this work, we describe an application of smartphones to monitor sleep apnea patients. Built-in sensors, such as microphones and accelerometers, are proposed to acquire information of snoring and respiratory sounds, and respiratory activity during the night. A study with polysomnography and smartphone system was completed at a sleep laboratory. The capability to detect sleep apnea episodes was validated by comparison of both systems and annotations by an expert in sleep medicine. The results of this study suggest that acoustic analysis of snoring and accelerometer signals, through smartphones, can be a feasible alternative to screen and monitor sleep apnea patients at home.

Keywords: Cardiovascular and respiratory system modeling - Sleep-cardiorespiratory Interactions, Sleep - Obstructive sleep apnea, Cardiovascular, respiratory, and sleep devices - Wearables
Abstract: A prompt diagnosis and treatment of OSA is vital for the healthy growth and development of many children. Polysomnography is costly and resource intensive resulting in long waitlists. There is a need to develop portable, easy-to-use, at-home tools to screen for OSA. Several studies have investigated the potential of overnight blood oxygen saturation (SpO2) analysis to provide a standalone OSA screening tool. In this study, to detect children with significant OSA, we propose an approach that involves a novel SpO2 visualization through a self-affine transformation, and feature extraction based on the distribution of the cloud points. The overnight oximetry (recorded using a smartphone-based pulse oximeter) of 190 children admitted to the British Columbia Children’s Hospital (BCCH) for a PSG was used to develop the screening tool. We trained an extreme learning machine (ELM) classifier using the extracted features to identify children with OSA, defined at different apnea/hypoapnea indices using a 10-fold cross-validation. The self-affine SpO2 visualization was more scattered in children with OSA (defined as AHI≥5) than in children without OSA (AHI<5). The proposed classification method using the features extracted from the SpO2 self-affine visualization was able to reliably discriminate children with and without OSA with an average AUC of 0.82 (±0.07). With further analysis and external validation, the proposed method can be developed as a robust OSA screening tool.

Keywords: Brain-computer/machine interface, Neural stimulation - Deep brain, Neurorehabilitation
Abstract: Brain-computer interfaces are realized with non-invasive and invasive sensors and allow to realize important medical applications: stroke rehabilitation, assessment and communication with patients with disorders of consciousness (DOC), avatar control, functional mapping of the eloquent cortex and deep brain stimulation in Parkinson patients.

Keywords: Neurological disorders, Neural signal processing, Neural stimulation - Deep brain
Abstract: Optimal placement of deep brain stimulation (DBS) electrode into the motor territory of the subthalamic nucleus (STN) is critical for the efficacy of stimulation and minimization of its side effects. Intraoperative processing of local field potentials (LFPs) recorded from microelectrodes can improve electrode placement into the target and overcome limitations in stereotaxic neuroimaging and single cell recordings. Here we present results obtained from 22 patients and show that spatio-spectral patterns of LFPs can be used effectively in clinical decision making for the accurate placement of chronic DBS electrodes.

Keywords: Neural signal processing, Brain physiology and modeling - Nonlinear coupling
Abstract: The ability to reveal important cortical regions supports neurosurgeons to optimize the functional outcome of brain surgeries. Electrical cortical stimulation (ECS) mapping turned out to be most reliable functional mapping technique so far, but comes with substantial downsides. Passive mapping using electrocorticography (ECoG) has been repetitively demonstrated to be clinically relevant, overcoming the risks of seizures and pain. Combining the strengths of both modalities opens new ways to reveal brain functions and at the same time reduces the patient’s compliance.

Keywords: Brain-computer/machine interface, Motor neuroprostheses - Neuromuscular stimulation, Neurological disorders - Treatment methodologies
Abstract: Brain Computer Interface (BCI) neuroprosthetics show promise for improving paralyzed patients’ functional independence by enabling thought-control of robotic arms or evoking movements in the patients’ own limbs. At present, BCI-neuroprosthetics are research devices, but we are a few critical advances away from being able to deploy them as assistive devices in patients’ homes. This mini- symposium topic will include recent clinical trials using BCI technology, an overview of the engineering systems linking neural activity to evoked movements, and effector systems that range from robotic arms to functional electrical stimulation (FES) of the users’ limbs. Leading researchers and an end-user will discuss their trials, innovations that allow improved neural decoding of fine motor control, sensory feedback, ways to improve neuroprosthetic training, barriers to deployment in patients’ homes, and expected technological advances that may help to address translational barriers.

Keywords: Brain-computer/machine interface, Motor neuroprostheses, Neurorehabilitation
Abstract: Brain computer interfaces (BCIs) have had several recent successful laboratory demonstrations, raising hopes that a take-home system could significantly improve the lives of patients in the near future. However, many challenges remain in translating BCI control of an assistive device in the lab into a robust take-home system. One challenge is designing neural decoders, the algorithms that translate neural activity into control commands for an assistive device, that meet patients’ performance expectations. Here, we review patient priorities for BCI systems from the literature to extract a set of requirements for the neural decoding component of BCI systems.

Keywords: Brain-computer/machine interface, Motor neuroprostheses - Neuromuscular stimulation
Abstract: Persons with chronic tetraplegia can regain function through the use of functional electrical stimulation and implanted brain recording microelectrodes. This mini-symposium talk details our work in restoring reaching and grasping to a person with chronic tetraplegia. We discuss hurdles and challenges to clinical translation of these systems.

Keywords:
Abstract: Brain computer interfaces (BCIs) have successfully been used in laboratory settings to restore upper limb motor function to individuals paralyzed from spinal cord injury. However, translation into neuroprosthetics for home use requires optimization informed by patient-centered design. Here, we review patient priorities from the literature and describe GAIN, a patient-centric framework for evaluating clinical performance of BCI-grasp neuroprosthetics.

Keywords: Brain-computer/machine interface, Motor neuroprostheses - Robotics, Neural interfaces - Microelectrode technology
Abstract: Brain-computer interfaces (BCIs) have the potential to restore upper limb function for people with tetraplegia, but few studies have quantified long term performance. We show that two people with tetraplegia could use an intracortical BCI to control a robotic arm with a high level of performance on reaching and grasping tasks over multiple years as measured with the Action Research Arm Test.

Keywords: Neurorehabilitation, Motor neuroprostheses - Neuromuscular stimulation, Neuromuscular systems - EMG processing and applications
Abstract: Brain-machine interfaces (BMIs) and myoelectric control interfaces (MCIs) can control functional electrical stimulation (FES) systems to replace or rehabilitate motor functions after spinal cord injury and stroke. Such systems are often complex and cumbersome, limiting their utility in community settings. This paper presents early results from work aimed at creating an easy-to-use system for FES-assisted hand grasp under myoelectric control that leverages a dense array of EMG sensors (HD-EMG) on the forearm.

Keywords: Health Informatics - eHealth, Health Informatics - Emerging IT for efficient/low-cost healthcare delivery, Health Informatics - Health data acquisition, transmission, management and visualization
Abstract: Abstract— Wireless medical systems have unique hazards and risks that need to be addressed in the concept, design, testing, and operation phases. These are especially important to consider and address in medical device systems where serious injury or death can occur related to the failure, disruption, or loss of information via wireless transmissions. This mini-symposium is focused on these issues with information about the technology trends, design and regulatory considerations, and practical advice that includes how to deal with the risks related to important aspects such as wireless coexistence and security. The session includes a survey of the trends for wireless medical devices over the last 20 years, information about the most widely used wireless technologies, considerations for digital health and the IoH (Internet of Health) and practical information that can help take the research concepts through some of the hurdles that lead to fruition and into deployment.

Keywords: Health Informatics - eHealth, Health Informatics - Emerging IT for efficient/low-cost healthcare delivery, Health Informatics - Health data acquisition, transmission, management and visualization
Abstract: Wireless medical systems have unique hazards and risks that need to be addressed in the concept, design, testing, and use. These are especially important to consider and address in medical device systems where serious injury or death can, and has, occurred related to the failure, disruption, or loss of information via wireless transmissions. This mini-symposium is focused on these issues with information about the trends, design and regulatory considerations, and practical advice that includes how to deal with the risks related to important aspects such as wireless coexistence and security. The session includes a survey of the trends for wireless medical devices over the last 20 years, information about the most widely used wireless technologies, considerations for digital health and the IoH (Internet of Health) and practical information that can help take the research concepts through some of the hurdles that lead to fruition and into deployment.

Keywords: Health Informatics - Quality of service, trust, security, Health Informatics - Health information systems, Health Informatics - Healthcare communication networks
Abstract: Wireless medical systems have unique hazards and risks that need to be addressed in the concept, design, testing, and use. These are especially important to consider and address in medical device systems where serious injury or death can, and has, occurred related to the failure, disruption, or loss of information via wireless transmissions. This mini-symposium is focused on these issues with information about the trends, design and regulatory considerations, and practical advice that includes how to deal with the risks related to important aspects such as wireless coexistence and security. The session includes a survey of the trends for wireless medical devices over the last 20 years, information about the most widely used wireless technologies, considerations for digital health and the IoH (Internet of Health) and practical information that can help take the research concepts through some of the hurdles that lead to fruition and into deployment.

Keywords: Sensor Informatics - Behavioral informatics, Sensor Informatics - Sensor-based mHealth applications
Abstract: Smartphones represent a major new opportunity for digital psychiatry: Used by more than a third of the world’s population, they are constant personal companions which process speech frequently as part of their most basic function, and can employ apps designed to elicit any desired types of speech. This presentation explores the opportunities and challenges for automatic detection and monitoring of mental state using smartphone-based speech signal processing approaches, with particular reference to depression.

Keywords: Health Informatics - e-communities, social networks and social media
Abstract: This investigation aims to determine whether individuals’ suicide risk can be profiled using their Twitter posts. Linguistic expression, and well as other users’ rates of reply, were investigated as markers of suicidality and the need for intervention.

Keywords: Health Informatics - Mobile health, General and theoretical informatics - Data intelligence, General and theoretical informatics - Computational phenotyping
Abstract: About 41% of individuals will experience depression at some point in their lives. One way to improve the care for this population is to monitor symptoms of patients over time. One emerging area is collecting data from smartphones and wearables and finding the relation to validated depression ratings. In this study, we collected short daily questionnaires, geolocation and cardiovascular information from a large cohort of bipolar and borderline personality disorder patients longitudinally and define metrics that are relevant for defining the state and severity of the disorder.

Keywords: Health Informatics - Mobile health, Sensor Informatics - Behavioral informatics, Sensor Informatics - Sensor-based mHealth applications
Abstract: In 2014, US spent a whopping 220 billion in providing mental health care services. Besides the substantial costs, the number of Americans (1 in 5) that suffer from some mental illness and its sequelae continue to rise. Timely access to cost-effective assessment, diagnosis, and treatment remain a crucial challenge. Smartphones and wearable technology have shown encouraging early trends in obtaining more accurate, momentary and contextual data about one's behavioral state outside the clinic. In this talk, I will provide a brief survey of the technology used in assessing major depressive disorder(MDD), including learnings from two large fully mobile based randomized control trials, our ongoing work in NIH's Precision Medicine Initiative (AllofUS), future opportunities, and challenges that lie ahead.

Keywords: Health Informatics - Mobile health, Health Informatics - Preventive health, Health Informatics - Patient tracking
Abstract: Suicide is a leading cause of death globally, but models of predicting risk are limited – a previous suicide attempt is the strongest single risk factor, however this has limited predictive ability at an individual level. Smartphone sensors are enabling a new class of behavioral markers to be developed, which can integrate with existing suicide prevention strategies to deliver a just-in-time intervention during a crisis. These opportunities, and methodological challenges, are discussed.

Keywords: Public Health Informatics - Epidemiology, Sensor Informatics - Wearable systems and sensors, Health Informatics - Technology and services for home care
Abstract: Social determinants of health such as local environments are well known risk factors as well as indicators of prognosis in mental health conditions. However, accurate reporting of the impact of environment on mental health has been limited by use of self-reported and retrospective data. In this 3-month observational study, 17 participants with schizophrenia utilized a smartphone app that recorded passive data including GPS. Using the ESRI ArcGIS 10.3 geospatial analysis platform, we compared time spent by participants with schizophrenia versus the general population in sensitive Environmental Justice (EJ) regions. We found that those with schizophrenia do spend more time in places with likely higher environmental burdens and stressors compared to the general population.

Keywords: Image reconstruction and enhancement - Machine learning / Deep learning approaches, Magnetic resonance imaging - Parallel MRI
Abstract: We introduce a model-based multi-channel MR image reconstruction framework with a convolution neural network (CNN) based regularization prior. The proposed formulation provides a systematic approach for deriving deep architectures for inverse problems with the arbitrary structure.

Keywords: Image reconstruction and enhancement - Machine learning / Deep learning approaches, Magnetic resonance imaging - Cardiac imaging, Regularized image Reconstruction
Abstract: Deep learning and artificial intelligence (AI) are re-inventing a number of aspects of medical imaging. Cardiac MRI has numerous challenges to obtain accurate functional and physiological properties in a short scan time. Undersampled image reconstructions using parallel imaging and compressed sensing are now standard practice for some types of cardiac imaging. Further improvement in acquisition efficiency and computational time may be possible with AI methods. This work gives a brief introduction to challenges in cardiac imaging and how AI methods may be re-inventing the approach to reconstructing such data.

Keywords: Magnetic resonance imaging - MR spectroscopy, Image reconstruction and enhancement - Machine learning / Deep learning approaches, Regularized image Reconstruction
Abstract: Magnetic resonance spectroscopic imaging (MRSI) has long been recognized as a potentially powerful tool for non-invasive, label-free molecular imaging. However, clinical and research applications of MRSI have been developing very slowly due to several long-standing technical barriers, including long data acquisition time, poor spatial resolution, low signal-to-noise (SNR), and overwhelming nuisance background signals. This talk will discuss our recent advances in addressing these challenging problems using a novel method that effectively integrates spin physics with machine learning.

Keywords: Novel imaging modalities
Abstract: Obesity is emerging as the major driver of the global non-communicable chronic disease epidemic. A challenge is to discover effect treatments that have enduring efficacy and high levels of safety. An important aspect of this search involves advances in telecommunications, internet, and related digital technologies that promise to become useful tools in the clinical management of obesity. This presentation will review these recent advances.

Keywords: Image registration, segmentation, compression and visualization - Volume rendering, Image segmentation, Image visualization
Abstract: Smartphone-based food photography methods rely on participants to capture images of their foods before and after eating, providing near real-time review of food intake. This is an effective tool to facilitate dietary change. Computer algorithms help automate image analysis, but most fail to function as hoped, and participants will not use food photography continuously for months or years. Energy balance models estimate long-term energy intake and adherence to energy intake targets, but do not provide data on what foods are eaten. The strengths of food photography and energy balance models can be combined to promote the efficacy of mHealth interventions.

Keywords: Image analysis and classification - Machine learning / Deep learning approaches, Optical imaging, Dual-energy X-ray imaging
Abstract: Malnutrition and increasingly sedentary lifestyles have led to an explosion in metabolic diseases worldwide. Body shape is an intuitive manifestation of metabolic status, and simple anthropometrics such as waist circumference and body mass index are known risk factors. However, these coarse metrics do not account for fat/lean composition or different body types. A technique to capture and analyze rich 3D body shape may offer stronger predictors of metabolic health risk across a wide range of body shapes and sizes. In this work, we investigate the use of whole body 3D photonic imaging and advanced statistical shape analysis to characterize metabolic risk and identify dominant human body phenotypes.

Keywords: X-ray radiography, Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, X-ray molecular imaging
Abstract: X-ray induced acoustic imaging has a great potential for providing X-ray absorption contrast with much reduced radiation exposure. In this paper, we have successfully demonstrated the feasibility of an X-ray induced acoustic imaging system using a portable X-ray source and a focused ultrasound transducer.

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, Novel imaging modalities, Optical imaging
Abstract: We report the method of cancer targeting using an squarine-based molecular probe and attenuated Salmonella typhimurium engineered to produce photoacoustic imaging signals. in tumor tissues. Engineered The engineered squarine molecules and bacteria can be imaged by photoacoustic imaging system.

Keywords: Optical imaging - Coherence tomography, Image feature extraction, Functional image analysis
Abstract: Tissue denaturation and coagulation process with thermal stress were monitored by speckle variance optical interferometry (SvOI) method. SvOI visualized 2D tomographic images of tissue denaturation and coagulation with microscale view. Furthermore, the state of the denatured biological tissue was calculated with the cross-correlation coefficient in the region of interest.

Keywords: Sensor Informatics - Behavioral informatics, Health Informatics - Behavioral health informatics, Sensor Informatics - Wearable systems and sensors
Abstract: Achievement of changes in eating behaviors toward those that facilitate long-term maintenance of weight loss is elusive. Emerging wearable sensor technology allows for accurate and objective measurement of ingestive behavior. Real-time analysis of the sensor data paves the way for development of individually tailored Just-In-Time Adaptive Interventions for weight control. This talk will cover an introduction to food intake detection using wearable sensors with a focus on our work in developing and using the Automatic Ingestion Monitor. Results from the studies on food intake detection, meal microstructure characterization and eating behavior modification will be reported.

Keywords: Imaging Informatics - Image analysis, processing and classification, Imaging Informatics - Image databases, Health Informatics - Mobile health
Abstract: One of the most important behavior both socially and culturally that impacts one’s health is diet. A measure of accurate dietary intake is considered to be an open research problem in the field of nutrition and health. We are developing image-based tools using deep learning techniques to automatically obtain an accurate estimate of energy/nutrients consumed in individuals’ diet.

Keywords: Health Informatics - Behavioral health informatics, Sensor Informatics - Multi-sensor data fusion, Sensor Informatics - Physiological monitoring
Abstract: To help people to improve their health-related behaviors with just-in-time interventions, we developed a new multi-level fusion inference algorithm using wearable sensor data. It is based on computational mechanistic models of acute stress dynamics and its effect on heart rate, heart rate variability, and electrodermal activity. The novel algorithm is designed to mitigate the effect of signal distortions arising from monitoring stress in real life with wearable sensors

Keywords: Sensor Informatics - Multi-sensor data fusion, Sensor Informatics - Behavioral informatics, Sensor Informatics - Sensor-based mHealth applications
Abstract: This paper highlights lessons learned from a four-year ambulatory study, developed to measure Sleep, Networks, Affect, Performance, Stress, and Health using Objective Techniques (SNAPSHOT), which was run in seven cohorts of college students (N=321), collecting continuous wearable and mobile phone data, typically for a month each. This paper overviews the objectives of this study, challenges faced, and some key findings focused on detecting sleep patterns and detecting and forecasting mood changes.

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic
Abstract: We present our recent progress on linear-array-based photoacoustic imaging systems. The systems have been used to image vascular structures in the human palm, foot, and breast. To improve the imaging result, we used advanced light illumination, acoustic detection, and image reconstruction methods. Our results clearly demonstrate the clinical translational impact of photoacoustic imaging.

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, Optical imaging, Multimodal imaging
Abstract: We present our second generation pulsed laser diode based compact photoacoustic tomography (PLD-PAT-G2) system. This system is ultra-compact with single-element ultrasound transducer augmented with acoustic reflector (USTR) and a portable pulsed diode laser. High-speed imaging is achieved at high pulse repetition rate laser excitation with multi-USTR detection. The performance of this system is demonstrated using deep-tissue imaging and small animal in vivo imaging.

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, Image reconstruction and enhancement - Machine learning / Deep learning approaches
Abstract: Traditional quantitative photoacoustic imaging (PAI) relies on ill-conditioned optical inversion. Recent studies have shown that methods that are exempt from optical inversion are very effective, strongly suggesting alternative ways to solving the long standing problem.

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, Optical imaging and microscopy - Microscopy, Optical imaging and microscopy - Optical vascular imaging
Abstract: We present our recent progress on a super-fast photoacoustic microscopy with an ultra-wide filed of view. By using a novel polygon mirror scanner that can be immersed in water, we have achieved a B-scan frame rate of 1000 Hz with a scanning range of at least 13 mm. Hemodynamic responses in the whole mouse brain and ear were imaged

Keywords: Ultrasound imaging - Photoacoustic/Optoacoustic/Thermoacoustic, Optical imaging and microscopy - Microscopy, Optical imaging and microscopy - Optical vascular imaging
Abstract: We present our latest progress on photoacoustic microscopy (PAM) for structural, functional and metabolic imaging in vivo. Integrating hardware and software innovations, multi-parametric PAM enables comprehensive characterization of microvascular structure (diameter, tortuosity and density), mechanical property (resistance, wall shear stress, reactivity and permeability), hemodynamics (blood perfusion, oxygenation and flow), and the related tissue oxygen extraction and metabolism. This enabling technique has found broad applications in neuroscience, cardiovascular, cancer research.

Keywords: Vascular mechanics and hemodynamics - Pulse wave velocity, Cardiovascular and respiratory signal processing - Pulse transit time, Cardiovascular and respiratory signal processing - Blood pressure measurement
Abstract: Pulse transit time (PTT) is the measured quantity in cuffless estimation of arterial blood pressure (BP). For noninvasive devices, PTT is generally obtained from skin surface measurements by transducers registering a mechanical displacement due to the arterial pressure pulse (eg. tonometer or cuff pulsations) or a blood volume signal from the peripheral microcirculation using, for example, photoplethysmography (PPG). The basis of the relation between BP and PTT is the pressure dependency of arterial stiffness which can be measured by pulse wave velocity (PWV). This study compares the BP-PWV sensitivity using BP postural changes with PTT measured by tonometry, cuff pulsations and PPG.

Keywords: Cardiovascular and respiratory signal processing - Pulse transit time, Cardiovascular and respiratory signal processing - Cardiovascular signal processing, Cardiovascular and respiratory signal processing - Blood pressure measurement
Abstract: Pulse Arrival Time (PAT) is often considered as an acceptable approximation of the Pulse Transit Time (PTT) for the measure of the cuffless blood pressure, (C)BP. PAT, however, includes part of the Pre Ejection Period, PEP. In cardiac patients, it may be expected that PEP be affected by the heart dysfunction thus making the CBP estimation less accurate. In this study we addressed this issue by comparing PAT, PTT and PEP collected in 14 patients and 8 controls. The obtained results put in evidence that in patients the individual contribution of PEP to PAT is much more variable than in controls and may even be longer than the corresponding PTT. These findings foster further investigations on the accuracy of the PAT-based measure of CBP in patients.

Keywords: Cardiovascular and respiratory signal processing - Blood pressure measurement, Cardiovascular, respiratory, and sleep devices - Sensors, Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: Routine monitoring of blood pressure during general anesthesia relies on intermittent measurements with a non-invasive brachial cuff inflated every two to five minutes. While all these patients are equipped by a fingertip pulse oximeter, the acquired optical signals currently only provide SpO2 estimates. Our running clinical trial (NCT02651558) presents the first-ever demonstration that the optical signals acquired by a fingertip pulse oximeter can also be exploited to continuously detect blood pressure changes. Results from the first 8 enrolled patients show that the Optical Blood Pressure Monitoring (oBPM) algorithms can detect rapid blood pressure changes occurring during anesthesia with 94% of accuracy. The proposed solution is expected to allow major improvements in the safety of anesthetized patient’s, allowing early detection of hemodynamic changes occurring in between two routine blood pressure measurements performed with brachial cuffs.

Keywords: Cardiovascular and respiratory signal processing - Blood pressure measurement, Vascular mechanics and hemodynamics - Pulse wave velocity, Cardiovascular and respiratory signal processing - Pulse transit time
Abstract: We present a unimodal arterial compliance probe for cuffless evaluation of blood pressure (BP) parameters using ARTSENS® technology. The proposed technique could evaluate BP parameters using simultaneously measured arterial dimensions, local pulse wave velocity (PWV), and their variations over the cardiac cycle, without any calibration procedure. An in-vivo study was conducted on five healthy subjects using a prototype system to validate the proposed approach. Local PWV and arterial dimension were evaluated from the carotid artery in a beat-by-beat manner. RMSE in carotid BP measurements was less than 10.5 mmHg, illustrating the reliability of the proposed calibration-free cuffless BP evaluation technique.

Keywords: Cardiovascular and respiratory signal processing - Pulse transit time
Abstract: This minisymposium contribution focuses on the design and implementation of a wrist-worn sensing system – SeismoWatch 2.0 – for measuring seismocardiogram (SCG) and photoplethysmogram (PPG) signals from a subject placing the system against the chest. The ultimate goal is to use these two measured signals to compute pulse transit time (PTT) and thereby estimate blood pressure (BP). The contribution represents the description of the second iteration of the watch, with PPG sensors both built into the inner part of the wrist strap (for measuring blood volume pulse at the radial artery) and on the outer face of the strap (for measuring the pulse on the skin surface at the sternum).

Keywords: Vascular mechanics and hemodynamics - Pulse wave velocity, Cardiovascular and respiratory signal processing - Blood pressure measurement, Cardiovascular and respiratory signal processing - Cardiovascular signal processing
Abstract: This study presents a novel robotic tonometry device to precisely measure and analyze the radial artery pulse waveforms. Because the radial artery lays under the skin layer, it is difficult for the operator to detect the exact pulsatile position. The proposed device consists of a six DOF robotic manipulator with a redundancy actuator and a passive universal joint with a linear spring component to find autonomously the pressurizing position and direction of the pulse sensor.

Keywords: Cardiovascular, respiratory, and sleep devices - Sensors, Cardiovascular and respiratory signal processing - Blood pressure measurement, Cardiovascular, respiratory, and sleep devices - Monitors
Abstract: This study analyzes the dynamic pulse wave characteristics of a fabricated pulse sensor with a cover layer according to the thickness and hardness of the cover materials and suggests an appropriate thickness and hardness to help people design a pulse sensor with a semiconductor device-based pressure sensor.

Keywords: Cardiovascular and respiratory signal processing - Non-linear cardiovascular or cardiorespiratory relations, Cardiovascular and respiratory system modeling - Cardiovascular Disease
Abstract: Definition of pulse patterns according to Korean medicine are uncertain and there are no gold standards. This study proposes a pulse pattern classification system based on fuzzy c-means clustering and fuzzy theory.

Keywords: Cardiovascular and respiratory system modeling - Blood flow models
Abstract: Oriental medicine students that are learning the technique of pulse diagnosis need a training tool that allows them to be able to practice feeling specific pulse signatures. Developing a mechanical simulator would allow students to have a practice tool, as well as set a standard for practitioners to refer to for diagnosis. In this paper, we introduce a compact radial pulsation simulator that uses a cam with three peak points created through mathematical transformation of human's pulse wave data. Experimental results confirmed that the developed cam system can regenerate human's pulse data very precisely.

Keywords: PET and SPECT imaging, Image reconstruction and enhancement - Machine learning / Deep learning approaches, Image enhancement - Denoising
Abstract: Recently, deep learning has outperformed the traditional machine learning and Bayesian approaches in many different applications such as image restoration and super-resolution with the large dataset and high computing power graphical processing unit. Furthermore, the deep learning approach apparently has great potential for providing answers to unsolved problems in medical imaging physics and engineering. In this talk, the challenges in radiation image generation and analysis and deep learning approaches to overcome these challenges will be presented.

Keywords: Magnetic resonance imaging - Diffusion tensor, diffusion weighted and diffusion spectrum imaging, Brain imaging and image analysis, Image analysis and classification - Machine learning / Deep learning approaches
Abstract: Deep learning, especially convolutional neural network (CNN) becomes a crucial methodology of analyzing medical images. However, few studies have done with the brain networks. We here compare the performance of CNN and a simple feedforward neural network using the brain network extracted from the diffusion-weighted MR images to investigate applicability of CNN to the brain networks.

Keywords: Neural interfaces - Implantable systems, Neural interfaces - Tissue-electrode interface, Neural interfaces - Microelectrode technology
Abstract: Neural interfaces have been shown to be effective for neuromodulation of sensory and cognitive systems as well as for recording of motor intention. Increasingly, cellular-level communication is desired, requiring higher-resolution interfaces with smaller electrodes that penetrate into the tissue. However, traditionally these interfaces have suffered from rejection in the body and unreliable long-term performance. Recently, manufacturing techniques and advanced materials have enabled a revolution in electrode design, enabling devices with architectures on the cellular scale (~<10 μm). This minisymposia will highlight examples and performance from novel subcellular neural interfaces that have been shown to result in a decreased biotic rejection and more stable long-term performance.

Keywords: Neural interfaces - Microelectrode technology, Neural interfaces - Tissue-electrode interface
Abstract: A long-term neural interface should have properties that match those of single axons. Here we focus on the mechanical properties of wire interface with diameters similar to that of axons. The flexural rigidity (FR) of platinum-iridium (Pt/Ir) wires and carbon nano-tube (CNT) yarns were measured and compared. The diameter (10µm) and flexural rigidity of the CNT yarn neural interface (3.3 10-12 N·m2) are similar to that of a single axon. This novel device provides the basis for a platform technology for long term recording of neural signals in small nerves such as those of the autonomic nervous system.

Keywords: Neural interfaces - Microelectrode technology
Abstract: The Paradromics Neuroscience Research System (NRS) is a multi-channel neural amplifier and data acquisition system for use with large arrays of microwire electrodes. It consists of a full signal chain: signals from the electrodes are amplified on a custom integrated circuit, digitized and readout on a rigid-flex printed circuit board, and recorded on a data acquisition server running custom high-performance interface software. The NRS allows recording from up to 65,536 simultaneous channels at rates up to 39 kHz with a resolution of 12-bits per digitized sample.

Keywords: Neural interfaces - Microelectrode technology, Neural interfaces - Tissue-electrode interface, Neural interfaces - Implantable systems
Abstract: Implanted electrodes provide one of the most important techniques for both basic and translational neuroscience by allowing for time-resolved electrical detection of neural activity in the living brain, but scalable and stable neural recording that can track a large ensemble of neurons from days to weeks and months remains challenging. Extensive efforts have been made to reduce the dimension and mechanical stiffness of neural electrodes for improved biocompatibility and recording reliability. Here, we present our recent progress on ultraflexible nanoelectronic threads (NETs) that drastically reduce the dimension and mechanical compliance of neural electrodes, which allow for seamless tissue integration and great promise of large-scale, long-term stable recording.

Keywords: Neural interfaces - Microelectrode technology, Neural interfaces - Tissue-electrode interface, Neural interfaces - Implantable systems
Abstract: Monitoring the electrical or chemical activity from neighboring but distinct populations of neurons while maintaining a stable interface is crucial to better understanding neural dynamics and information processing. To achieve this, our group has developed two high density carbon fiber arrays with the ability to monitor electrophysiology or dopaminergic activity. One design has been implanted in the nucleus accumbens of rats with successful recordings from both modalities at least two months post-implantation. Due to the size of the electrodes, it is possible to leave the implants in place during tissue processing, allowing for a better and unperturbed understanding of local circuit dynamics

Keywords: Biologically inspired robotics and micro-biorobotics - Modeling, Biomimetic robotics, Micro-and nano-biorobotics
Abstract: This paper describes a design concept and a prototype development of Bionic Humanoid for medical simulator and training in eye surgery and neurosurgery. Based on the concept of Bionic Humanoid, we established a brand-new eye surgery simulator, Bionic Eye surgery Evaluator (Bionic-EyE), for training of peeling the inner limited membrane (ILM) which is superficial layer of retina. Recent progress of Bionic Humanoid will be introduced and discussed for future medical innovation.

Keywords: New technologies and methodologies in biomechanics, Applied tissue and organ models and motion analysis
Abstract: Abstract— This paper describes how to finalize a specification of clinical version of implantable left ventricular assist system, determined from the medical regulatory science. Our benefit/risk analysis indicated that the controller, battery and cool-seal unit should be located outside of the body, because we can detect an emergency event which occurred inside the pump immediately.

Keywords: Prosthetics - Robotic organs, New technologies and methodologies in medical robotics, Robot-aided surgery - Targeted therapy
Abstract: This paper describes a novel fully implantable artificial pancreas concept. A non-invasive insulin refilling strategy, based on ingestible smart capsules and a non-invasive battery recharging technology, based on non-radiative transcutaneous power transfer, pave the way towards the development of a potentially revolutionary paradigm for type 1 diabetes treatment. This has the potential to allow diabetic patients to almost forget their pathology.

Keywords: Prosthetics - Robotic organs, Robotic prosthetics
Abstract: This paper describes novel components for an artificial urinary system, namely an artificial bladder and an artificial sphincter. An accurate analysis of the best shape of the bladder has been performed, together with a selection of internal coating materials guaranteeing bladder stability. Magnetic activation has been selected for the artificial sphincter, thus allowing to eliminate the implantation of active components and batteries. Preliminary prototypes and tests are presented, towards the realization of a biomechatronic urinary system.

Keywords: Prosthetics - Robotic organs, Biomimetic robotics, Prosthetics - Bionic sensory systems
Abstract: Fully implantable biomechatronic organs deal with the set of implantable technologies that establish a close relationship between the patients’ biological structures and the artificial machine components, with the aim to restore some physiological functions that have been lost or severely compromised. These systems are also defined as “bionic organs”. In this paper, the disciplines that can contribute to the improvement of bionic organs, the open challenges and possible future directions in this field are discussed.

Keywords: New technologies and methodologies in medical robotics
Abstract: Implantable soft robotic devices are attractive due to their atraumatic nature. Here we present a soft active sleeve employed as a cardiac ventricular assist device that can deliver mechanical assistance to an arrested or failing heart, without contacting blood. Our approach is to employ a biologically inspired design, whereby individual contracting actuators are oriented in a layered helical and circumferential fashion, mimicking the orientation of the epicardial and myocardial fibers of the heart. We introduce methods for fabricating such an active heart sleeve, and show the effect of different control schemes on device performance in vitro and ex vivo.

Keywords: Optical imaging - Coherence tomography, Ophthalmic imaging and analysis, Image segmentation
Abstract: Polarization sensitive optical coherence tomography provides an additional contrast channel that enables imaging of ocular structures with intrinsic, tissue specific contrast. The method and examples of imaging in healthy and diseased eyes are presented.

Keywords: Ophthalmic imaging and analysis, Optical imaging - Coherence tomography
Abstract: We developed an attention detection and polarization sensitive central-fixation monitoring system to be used in combination with an optical coherence tomography (OCT) system, to facilitate retinal imaging in young children. The system integrates three major components: a) a Retinal Birefringence Scanning (RBS) subsystem that detects central fixation by locating the position of the fovea, b) a computer-controlled video player that plays attention attracting movies and directs the subject’s fixation to a central point target, and c) an optical coherence tomography subsystem for acquiring 3D images from the retina. The RBS subsystem, in a double-pass configuration, scans continuously the area around the fovea using polarized light and a fast circular scan, and analyzes the changes of polarization caused by the Henle fibers around the fovea. This allows fast detection of central fixation. The OCT system is directed by the RBS software to acquire and/or analyze data if and only if central fixation is detected, thus avoiding the analysis of large amounts of irrelevant data, and speeding up the overall performance.

Keywords: Ophthalmic imaging and analysis, Novel imaging modalities, Optical imaging
Abstract: Corneal birefringence is a well-known confounding factor and must be dealt with in polarization-sensitive technology used for retinal scanning and other intraocular assessment. For applications that are geared towards children, an approach of bypassing rather than compensating for the corneal birefringence, as described here, is desired.

Keywords: Ophthalmic imaging and analysis
Abstract: A half wave plate spinning at the proper frequency, plus signal processing using phase shift subtraction, can provide differential measurement of a periodic polarization signal using a single detector. Choice of the amount of phase shift can eliminate noise frequencies and enhance desired frequencies. This is successfully applied to an ophthalmic retinal scanning technique that we have devised and tested. Our signal is periodic, arising from scanning a spot of linearly polarized, near-infrared light in a 3-degree circle on the retina of a human eye. The goal is to sense foveal fixation of the eye via double-pass detection of the unique polarization signature of the radially arrayed birefringent Henle nerve fibers emanating from the fovea.

Keywords: Optical imaging - Coherence tomography, Optical imaging and microscopy - Optical vascular imaging
Abstract: This presentation describes basic principle and applications of Jones matrix optical coherence tomography (JM-OCT), which simultaneously visualizes scattering, polarization, and flow property of living tissue. In addition, an extension of JM-OCT, so called JM-optical coherence elastography is demonstrated. It visualizes not only the optical property of the tissue, but also its mechanical property.

Keywords: Computer model-based assessments for regulatory submissions, Computer modeling for treatment planning
Abstract: Four prototypes of high resolution surface-based human head models were constructed for simulation in a reasonable time by a commercially available frequency domain solver—ANSYS HFSS. The influence of cerebrospinal fluid and the upper part of the human torso on field propagation estimates of traveling wave excitation were evaluated at 297.2 MHz.

Keywords: Health technology - Verification and validation, Computer model-based assessments for regulatory submissions
Abstract: Design tasks in a variety of modern biomedical applications can be greatly accelerated through the use of numerical simulations. While a number of computational phantoms describing the human body exist, most are used under the assumption that they are accurate and produce representative results for the estimation of responses due to diverse electromagnetic stimuli. This study provides a framework for verification of a computational phantom and documents its use on the VHP-Female v.5.0 CAD model. The methodology presented herein may be applicable to any computational phantom given the availability of the source data.

Keywords: Computer model-based assessments for regulatory submissions, Pacemakers (implantable or external), Defibrillators (implantable or external)
Abstract: Despite its import as a diagnostic tool, patients with active implantable medical devices (AIMDs) are generally denied access to magnetic resonance imaging (MRI). The complexity of MRI environments stems from a multiplicity of fields and numerous scan parameters. In order to perform a risk assessment for RF-induced malfunction, manufacturers perform electromagnetic simulations using computational human models (CHMs). This work explores the impact of the CHM on the calculation of RF-induced voltages at the RF antenna port for cardiovascular implantable electronic devices (CIEDs).

Keywords: Neural stimulation (including deep brain stimulation), Neuromodulation devices
Abstract: A completely new type of integrated multichannel TMS/MRI imaging/stimulation system that enables unprecedented spatiotemporal control of the TMS-induced electric fields (E-fields) and simultaneous rapid whole-head MRI acquisition will be designed. Effects of the TMS elements on the B1+ field of the birdcage coil were investigated.

Keywords: Neural stimulation (including deep brain stimulation), Neuromodulation devices, Medical devices interfacing with the brain or nerves
Abstract: SimNIBS 2 (Simulation of Non-Invasive Brain Stimulation, v. 2) is a software pipeline for individualized electric field calculations for Transcranial Magnetic and Electric Stimulation (TMS, TES). We give an overview of the practical steps of the SimNIBS workflow from individualized head model generation to simulation of brain stimulation protocols, and finally the statistical analysis of the resulting electric fields in a subject group.

Keywords: Computer model-based assessments for regulatory submissions, Health technology - Verification and validation, Computer modeling for treatment planning
Abstract: The interaction between the RF field encountered during an MRI scan and an implanted medical device can cause heating in surrounding tissues. Computational modeling and simulation (CM&S) tools provide a platform for evaluating the MRI safety of implanted devices, but model credibility remains a challenge. This talk reviews the current verification and validation (V&V) best practices for establishing the credibility of a computational model for MRI safety labeling.

Keywords: Brain-computer/machine interface, Neural signals - Blind source separation (PCA, ICA, etc.), Neural signal processing
Abstract: We are developing an autonomously updating brain machine interface (BMI) utilizing reinforcement learning principles. One aspect of this system is a neural critic that determines reward expectations from neural activity. This critic is then used to update a BMI decoder toward an improved performance from the user’s perspective. Here we demonstrate the ability of a neural critic to classify trial reward value given activity from the primary motor cortex (M1), using neural features from single/multi units (SU/MU), and local field potentials (LFPs) with prediction accuracies up to 97% correct. A nonhuman primate subject conducted a cued center out reaching task, either manually, or observationally. The cue indicated the reward value of a trial. Features such as power spectral density (PSD) of the LFPs and spike-field coherence (SFC) between SU/MU and corresponding LFPs were calculated and used as inputs to several classifiers. We conclude that hybrid features of PSD and SFC show higher classification performance than PSD or SFC alone (accuracy was 92% for manual tasks, and 97% for observational). In the future, we will employ these hybrid features toward our autonomously updating BMI.

Keywords: Brain-computer/machine interface, Sensory neuroprostheses - Auditory, Neural signal processing
Abstract: Recent technological advances in the design of concealable miniature electroencephalography (mini-EEG) devices are paving the way towards 24/7 neuromonitoring applications in daily life. However, such mini-EEG devices only cover a small area and record EEG over much shorter inter-electrode distances than in traditional EEG headsets. These drawbacks can potentially be compensated for by deploying a multitude of such mini-EEG devices and then jointly processing their recorded EEG signals. In this study, we simulate and investigate the effect of using such multi-node EEG recordings in which the nodes are galvanically separated from each other, and only use their internal electrodes to make short-distance EEG recordings. We focus on a use-case in auditory attention detection (AAD), and we demonstrate that the AAD performance using galvanically separated short-distance EEG measurements is comparable to using an equal number of long-distance EEG measurements if in both cases the electrodes are optimally placed on the scalp. To this end, we use a group-LASSO based channel selection method in order to find these optimal locations.

Keywords: Brain-computer/machine interface, Human performance - Gait, Brain functional imaging - EEG
Abstract: We outline an experimental setup designed to dynamically understand neural responses to visual cliffs while walking. The goal of our work is understanding and mitigating fear of falling, particularly among the elderly. In our setup, an EEG cap monitors a subject's neural activity while the subject is immersed in a virtual world and walking on an instrumented treadmill. The subject's response to visual stimuli is measured by both the EEG cap and by speed and pressure data from the treadmill. Based on this data, we can dynamically alter the landscape in the virtual world. We hope that our setup may be useful in helping subjects develop mechanisms to compensate for significant fear of falling while walking.

Keywords: Brain-computer/machine interface, Neural signal processing, Brain functional imaging - NIR
Abstract: Near Infrared Spectroscopy (NIRS)-based Brain Computer Interfaces (NIRS-BCI) rely mainly on the mean concentration changes and slope of the hemodynamic responses in separate recording channels to detect the mental-task related brain activity. Nevertheless, spatial patterns across the measurement channels are also present and should be taken into account for reliable evaluation of the aforementioned detection. In this work the Dirichlet Energy of NIRS signals over a graph is considered for the definition of a measure that would take into account the spatial NIRS features and would integrate the activity of multiple NIRS channels for robust mental task related activity detection. The application of the proposed measure on a real NIRS dataset demonstrates the efficiency of the proposed measure.

Keywords: Brain-computer/machine interface, Neural signal processing, Brain functional imaging - EEG
Abstract: Recent studies have shown that using the user’s average steady-state visual evoked responses (SSVEPs) as the template to template-matching methods could significantly improve the accuracy and speed of the SSVEP-based brain-computer interface (BCI). However, collecting the pilot data for each individual can be time-consuming. To resolve this practical issue, this study aims to explore the feasibility of leveraging prerecorded datasets from the same users by transferring common electroencephalogram (EEG) responses across different sessions with the same or different electrode montages. The proposed method employs spatial filtering techniques including response averaging, canonical correlation analysis (CCA), and task-related component analysis (TRCA) to project scalp EEG recordings onto a shared response domain. The transferability was evaluated by using 40-class SSVEPs recorded from eight subjects with nine electrodes on two different days. Three subsets of electrode montages were selected to simulate different scenarios such as identical, partly overlapped, and non-overlapped electrode placements across two sessions. The target identification accuracy of the proposed methods with transferred training data significantly outperformed a conventional training-free algorithm. The result suggests training data required in the BCI speller could be transferred from different EEG montages and/or headsets.

Keywords: Brain-computer/machine interface, Motor neuroprostheses, Neural interfaces - Implantable systems
Abstract: Neural prostheses are being developed to restore speech to people with neurological injury or disease. A key design consideration is where and how to access neural correlates of intended speech. Most prior work has examined cortical field potentials at a coarse resolution using electroencephalography (EEG) or medium resolution using electrocorticography (ECoG). The few studies of speech with single-neuron resolution recorded from ventral areas known to be part of the speech network. Here, we recorded from two 96-electrode arrays chronically implanted into the 'hand knob' area of motor cortex while a person with tetraplegia spoke. Despite being located in an area previously demonstrated to modulate during attempted arm movements, many electrodes' neuronal firing rates responded to speech production. In offline analyses, we could classify which of 9 phonemes (plus silence) was spoken with 81% single-trial accuracy using a combination of spike rate and local field potential (LFP) power. This suggests that high-fidelity speech prostheses may be possible using large-scale intracortical recordings in motor cortical areas involved in controlling speech articulators.

Keywords: Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification, Time-frequency and time-scale analysis - Nonstationary processing
Abstract: The need of a reliable drowsiness detection system is arising today, as drowsiness is considered as a major cause for accidents as much as alcohol. In this paper, we propose a real-time drowsiness detection algorithm based on a single-channel electroencephalography (EEG) for wearable devices without demanding computing and power resources. The proposed algorithm adopts a cumulative counter to extract important features from 8 different frequency bands: delta (1–3 Hz), theta (4–7 Hz), low-alpha (8–9 Hz), high-alpha (10–12 Hz), low-beta (13–17 Hz), high-beta (18–30 Hz), low-gamma (31–40 Hz), and high-gamma (41–50 Hz). These features are then processed by a support vector machine (SVM) to distinguish between drowsy and awake states. Our preliminary results demonstrate that the proposed algorithm is capable of detecting drowsiness with superior accuracy (83.36%) over the conventional method (70.62%).

Keywords: Signal pattern classification, Connectivity measurements, Data mining and processing - Pattern recognition
Abstract: Power and connectivity features extracted from EEG signals have been previously utilized to detect mental fatigue during driving. Although each of the feature categories has the discriminative power to differentiate alert and fatigue states, they might represent different aspects of information relevant to fatigue identification. Two fusion methods, feature level and decision level fusions, were proposed in this study to combine individual channel information (i.e., power features) and between-channel information (i.e., connectivity features). According to the results of the study, the average accuracies of the fusion methods were higher than the accuracies of the individual feature categories (feature level fusion: 84.70%, decision level fusion: 87.13%, power features: 80.82%, connectivity features: 79.36%). The statistical analysis demonstrated that the two fusion methods significantly improved the classification performance of driving fatigue. The fusion methods proposed in this study can be embedded into a driving fatigue detection system for a practical use in a vehicle.

Keywords: Signal pattern classification, Independent component analysis
Abstract: Non-brain contributions to electroencephalographic (EEG) signals, often referred to as artifacts, can hamper the analysis of scalp EEG recordings. This is especially true when artifacts have large amplitudes (e.g., movement artifacts), or occur continuously (like eye-movement artifacts). Offline automated pipelines can detect and reduce artifact in EEG data, but no good solution exists for online processing of EEG data in near real time. Here, we propose the combined use of online artifact subspace reconstruction (ASR) to remove large amplitude transients, and online recursive independent component analysis (ORICA) combined with an independent component (IC) classifier to compute, classify, and remove artifact ICs. We demonstrate the efficacy of the proposed pipeline using 2 EEG recordings containing series of (1) movement and muscle artifacts, and (2) cued blinks and saccades. This pipeline is freely available in the Real-time EEG Source-mapping Toolbox (REST) for MATLAB (The Mathworks, Inc.).

Keywords: Physiological systems modeling - Closed loop systems, Physiological systems modeling - Multivariate signal processing, Signal pattern classification
Abstract: In the last decades numerous researches have revealed a strong link between emotions and several physiological responses. However, the automatic recognition of emotions still remains a challenge. In this work we describe a novel approach to estimate valence, arousal and dominance values from various biological parameters (derived from electrodermal activity, heart rate variability signal and electroencephalography), by means of multiple linear regression models. The models training was performed by using a set of pictures pre-evaluated in terms of valence, arousal and dominance, selected from the International Affective Picture System (IAPS) database. By using the step-wise regression method, all the possible combinations of considered biological parameters were tested as input variables for the models. The three multiple linear regression models that could provide the best fit for IAPS pictures valence, arousal and dominance values were selected. The features included in the optimal models were the average of the inter-beat duration (mean RR), the EEG spectral power computed in alpha, beta and theta frequency bands (Alpha, Beta and Theta power) and the average value of EDA signal (mean EDA). The obtained models show an overall good performance in predicting valence, arousal and dominance values.

Keywords: Connectivity measurements, Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification
Abstract: Brain connectivity measurements can provide key information about ongoing brain processes. In this paper, we propose to investigate the performance of the binary classification of Propofol-induced sedation states using partial granger causality analysis. Based on the brain connectivity measurements obtained from EEG signals in a database that contains four sedation states: baseline, mild, moderate, and recovery, we consider eight sensors and evaluate the area under the ROC curve with five classifiers: the k-nearest neighbor (density method), support vector machine, linear discriminant analysis, Bayesian discriminant analysis, and a model based on extreme learning machine. The results support the conclusion that the different Propofol-induced sedation states can be identified with an AUC of around 0.75, by considering signal segments of only 4 second. These results highlight the discriminant power that can be obtained from scalp level connectivity measures for online brain monitoring.

Keywords: Parametric filtering and estimation, Physiological systems modeling - Signal processing in physiological systems, Signal pattern classification
Abstract: Electroencephalogram (EEG) signals have been shown very effective for inferring user intents in brain-computer interface (BCI) applications. However, existing EEG-based BCIs, in many cases, lack sufficient performance due to utilizing classifiers that operate on EEG signals induced by individual trials. While many factors influence the classification performance, an important aspect that is often ignored is the temporal dependency of these trial-EEG signals, in some cases impacted by interference of brain responses to consecutive target and non-target trials. In this study, the EEG signals are analyzed in a parametric sequence-based fashion, which considers all trials that induce brain responses in a rapid-sequence fashion, including a mixture of consecutive target and non-target trials. EEG signals are described as a linear combination of time-shifted cortical source activities plus measurement noise. Using a superposition of time invariant with an auto-regressive (AR) process, EEG signals are treated as a linear combination of a stationary Gaussian process and time-locked impulse responses to the stimulus (input events) onsets. The model performance is assessed in the framework of a rapid serial visualization presentation (RSVP) based typing task for three healthy subjects across two sessions. Signal modeling in this fashion yields promising performance outcomes considering a single EEG channel to estimate the user intent.

Keywords: Brain imaging and image analysis, Image analysis and classification - Machine learning / Deep learning approaches
Abstract: Brain morphometry study plays a fundamental role in neuroimaging research. In this work, we propose a novel method for brain surface morphometry analysis based on surface foliation theory. Given brain cortical surfaces with automatically extracted landmark curves, we first construct finite foliations on surfaces. A set of admissible curves and a height parameter for each loop are provided by users. The admissible curves cut the surface into a set of pairs of pants. A pants decomposition graph is then constructed. Strebel differential is obtained by computing a unique harmonic map from surface to pants decomposition graph. The critical trajectories of Strebel differential decompose the surface into topological cylinders. After conformally mapping those topological cylinders to standard cylinders, parameters of standard cylinders (height, circumference) are intrinsic geometric features of the original cortical surfaces and thus can be used for morphometry analysis purpose. In this work, we propose a set of novel surface features rooted in surface foliation theory. To the best of our knowledge, this is the first work to make use of surface foliation theory for brain morphometry analysis. The features we computed are intrinsic and informative. The proposed method is rigorous, geometric, and automatic. Experimental results on classifying brain cortical surfaces between patients with Alzheimer's disease and healthy control subjects demonstrate the efficiency and efficacy of our method.

Keywords: PET and SPECT imaging, Brain imaging and image analysis, Magnetic resonance imaging - MR neuroimaging
Abstract: Prediction of Alzheimer’s disease (AD) from Mild Cognitive Impairment (MCI) by analyzing Magnetic Resonance Imaging (MRI) image features has become popular in recent years. However, defining effective predictive biomarkers is still challengeable. The ‘radiomics’ is a new method to identify advanced and high order quantitative imaging features and has been applied into oncology study. However, it has not been applied into brain disorder disease study. Therefore, the purpose of this study is to identify whether the ‘radiomics’ technique could be used to identify predictors of the conversion from MCI to AD. We analyzed 197 samples with MRI scans from the ADNI database, which contained 32 healthy subjects and 165 MCI patients. Firstly, we extracted 215 radiomics features from hippocampus. Then we used Cronbach’s alpha coefficient, the intra-class correlation coefficient, Kaplan-Meier model and cox regression to select 44 radiomics features as effective features. Finally, we used SVM classification to validate these features. The results showed that the classification accuracy using linear, polynomial and sigmoid kernel could achieve 80.0%, 93.3% and 86.6% to distinguish MCI-to-AD fast and slow converter. As a result, this study indicated that the ‘radiomics’ method is potential to be applied into predicting AD from MCI.

Keywords: Image segmentation, Brain imaging and image analysis
Abstract: Fast and accurate midline shift (MLS) estimation has a significant impact on diagnosis and treatment of patients with Traumatic Brain Injury (TBI). In this paper, we propose an automated method to calculate the amount of shift in the midline structure of TBI patients. The MLS values were annotated by a neuroradiologist. We first select a number of slices among all the slices in a CT scan based on metadata as well as information extracted from the images. After the slice selection, we propose an efficient segmentation technique to detect the ventricles. We use the ventricular geometric patterns to calculate the actual midline and also anatomical information to detect the ideal midline. The distance between theses two lines is used as an estimate of MLS. The proposed methods are applied on a TBI dataset where they show a significant improvement of the the proposed method upon existing approach.

Keywords: Brain imaging and image analysis, Image analysis and classification - Machine learning / Deep learning approaches, Image feature extraction
Abstract: Characterizing the cellular architecture (cytoarchitecture) of tissues in the nervous system is critical for modeling disease progression, defining boundaries between brain regions, and informing models of neural information processing. Extracting this information from anatomical data requires the expertise of trained neuroanatomists, and is a challenging task for inexperienced analysts. To address this need, we present an unbiased, automated method to estimate cellular density of retinal and neocortical datasets. Our approach leverages the fact that within retinal and neurocortical datasets, cells are organized into "layers" of constant density to approximate cytoarchitecture with a small number of known basis elements. We introduce methods for patch extraction, cell detection, and sparse approximation of inhomogeneous Poisson processes to differentiate changes in cellular densities and detect layers. Our results demonstrate the feasibility of using automation to reveal the cytoarchitecture of large-scale biological samples.

Keywords: Image analysis and classification - Machine learning / Deep learning approaches, Brain imaging and image analysis, Image classification
Abstract: Chordoma is a rare primary malignant tumor. For evaluating the related factors of postoperative recurrence probability of chordoma before surgery, we retrospective collected 80 patients to analyze by using a novel radiomics method. A total of 620 3D imaging features used for radiomics analysis were extracted, and 5 features were selected from T2-weighted (T2-w) magnetic resonance imaging (MRI) that were most strongly associated with 4-year recurrence probability to build a radiomics signature. Verification by logistic regression classification model, the area under the receiver operating characteristic curve and accuracy was 0.8600 (95% CI: 0.7226-0.9824) and 85.00% in the training cohort, respectively, while in the validation cohort was 0.8568 (95% CI: 0.7327-0.9758) and 85.00%. Experimental results show that T2-w MRI-based radiomics signature is closely associated with the recurrence of chordoma. It is possible to prejudge the recurrence of chordoma before surgery.

Keywords: Image analysis and classification - Machine learning / Deep learning approaches, Brain imaging and image analysis, Image registration, segmentation, compression and visualization - Volume rendering
Abstract: Whole mouse brain microvascular images at submicrometerscalecanbeobtainedbyKnife-EdgeScanning Microscopy (KESM). However, due to the large size of the image dataset and the noise from the serial sectioning process of the KESM, whole mouse brain vascular reconstruction and analysis with submicrometer resolution have not been achieved yet, while several previous studies demonstrated manually selected small noise-free portion of the KESM dataset. In addition to the KESM dataset, there have been studies for vessel reconstruction and analysis of the whole mouse brain at lower resolution or of partial brain regions at submicrometer resolution. Also, to get the brain contours, the authors performed manual editing. However, to the best of our knowledge, there has been no study for vessel reconstruction and analysis of the whole mouse brain at submicrometer resolution. In this paper, we propose a framework for the 3D reconstruction and analysis of the whole KESM mouse brain vasculature dataset with rich vasculature information extracted at submicrometer resolution. The framework consists of two methods. The propose methods provide the systematic cleaning to remove and consolidate erroneous images automatically, which enables the full tracing and analysis of the whole KESM mouse brain dataset with richer vasculature information.

Keywords: Sleep - Obstructive sleep apnea, Sleep - Cardiovascular & Metabolic consequences of sleep disorders, Cardiovascular and respiratory system modeling - Sleep-cardiorespiratory Interactions
Abstract: Sleep disordered breathing is a very common health problem with prevalence of 17% in men and 9% in women. Prevalence is increasing. In some patients sleep apnea may be a sign of the aging respiratory control and in others sleep apnea may be due to increased upper airway collapsibility, due to obesity, due to impaired respiratory control, or secondary due to cardiac or metabolic disorders. The identification of phenotypes and mortality markers is the new challenge. New biomarkers are necessary to separate groups of apnea patients in terms of phenotypes. Although home sleep studies become more and more common for the diagnosis of sleep apnea, the diagnostic reference remains to be polysomnography including a sleep EEG in a sleep laboratory setting. The sleep EEG is information rich and may serve as a source for new and reliable biomarkers to distinguish patients with different types and severities of sleep apnea. Possible markers are EEG patterns like sleep spindles, arousal, or statistical descriptors like number of transitions between sleep stages and transition probabilities between specific sleep stages.

New sleep EEG based biomarkers may serve as predictors for other sleep disorders such as insomnia, narcolepsy, or periodic leg movement syndrome. They may also serve as predictors for direct performance markers for daytime sleepiness.

Keywords: Sleep - Periodic breathing & central apnea, Sleep - Obstructive sleep apnea, Sleep - Cardiovascular & Metabolic consequences of sleep disorders
Abstract: Long-term sleep monitoring can be done primarily in the home environment. Good patient acceptance requires low user and installation barriers. The selection of parameters in this approach is significantly limited compared to a PSG session. The aim is a qualified selection of parameters, which on the one hand allow a sufficiently good classification of sleep phases and on the other hand can be detected by non-invasive methods.

Keywords: Sleep - Obstructive sleep apnea, Sleep - Sleep apnea therapy
Abstract: While it is well known that obstructive sleep apnoea (OSA) leads to chronic periodic blood oxygen desaturation during sleep, we have also found an association between moderate-severe OSA and decreased cerebral blood flow when awake. Currently, in a follow-up study, we are investigating the effect of OSA and CPAP treatment on brain blood flow, microsleep risk, and cognition, in people with moderate OSA.

Keywords: Sleep - Periodic breathing & central apnea, Cardiovascular, respiratory, and sleep devices - Diagnostics
Abstract: Sleep disordered breathing (SDB) with Cheyne Stokes Respiration (CSR) is often diagnosed in patients with chronic heart failure (CHF). CSR diplays a periodic breathing pattern with a typical waxing and waning breathing with central sleep apnea phases in between. Cycle length of CSR seems to be dependent of left ventricular ejection fraction which is a major predictor for death in patients with heart failure.

Keywords: Neural interfaces - Bioelectric sensors, Neural stimulation, Neural interfaces - Implantable systems
Abstract: Millions of people in the United States suffer from bladder dysfunction and bladder pain conditions. The underlying pathophysiology for many of these conditions is poorly understood. Studies seeking insights from animal models have been hindered by the inability to monitor bladder function in awake, freely behaving animals. To overcome this technical hurdle, we developed a low modulus wireless strain gauge that enables real-time measurement of dynamic changes in bladder size. This device also includes integrated microscale light emitting diodes (μLEDs) to allow optogenetic manipulation of opsins expressed in bladder sensory neurons. The strain gauge and μLEDs operate via an implanted Bluetooth base station that allows for wireless monitoring of bladder activity, and control of the μLEDs. We show this device can chronically measure and optogenetically modulate bladder activity in rats.

Keywords: Neural stimulation, Neural interfaces - Bioelectric sensors, Brain functional imaging - fMRI
Abstract: Despite widespread clinical use, the therapeutic results of vagal nerve stimulation (VNS) are often modest and inconsistent from patient to patient. Here Dr. Kip Ludwig of the Mayo Clinic will describe his team’s recent efforts to comprehensively characterize VNS in the swine model by incorporating intrafascicular nerve recordings, electromyography (EMG), high-frequency ultrasound, functional magnetic resonance imaging, as well as electrophysiological and neurochemical recordings in the deep brain. The goal is to understand how electrode location, configuration and stimulus parameters alter evoked fiber activity within the cervical vagus in the animal model most analogous to human patients, and study how these parameters can be optimized to minimize unintended VNS evoked activity within the neck muscles that limit dosing while maximizing intended therapeutic changes at the end-organ and in the brain.

Keywords: Neural stimulation
Abstract: This talk will focus on the use of HF10 therapy, a novel form of spinal cord stimulation (SCS) at 10 kHz for the treatment of focal, neuropathic pain from peripheral nerve injury. Pre-clinical studies conducted in a rodent model may shine light on the potential mechanism of action (MoA).

Keywords: Neural stimulation
Abstract: With the increase in interest in electrical stimulation of peripheral nerves, an emphasis on the anatomic and functional mapping of end-point structures within the ganglion-nerve unit is needed more than ever. The development of topographic mapping of postganglionic cell bodies within the ganglion will allow for identification of the precise electrical intervention needed for treatment of pathologic diseases. The Superior Cervical Ganglion (SCG) is chosen here as it projects to multiple head and neck structures. The focus of this study is to utilize a novel simple method of optical clearing technique to begin characterizing anatomic topographic mapping of the SCG to inform where functional pathways may exist in 3-D fashion. Here we demonstrate identification of ganglionic cells within the SCG from the nasopharynx which lead to identification of the functional pathways within the SCG.

Keywords: Sensory neuroprostheses, Motor neuroprostheses, Neural stimulation
Abstract: Optogenetics affords us a means to interface with the parasympathetic nervous system in a non-invasive manner by placing a miniature multiphoton microscope on the epineurium of the nerve and allowing us to image into the nerve. Neurons transfected to express an appropriate optical protein can be read-in or read-out. We have been able to elicit an optical response in the neuron at the nodes-of-Ranvier in response electrical stimulation of the nerve as well as genetically transfect neurons with GCaMP6f using an Adeno Associated Virus (AAV).

Keywords: Neural interfaces - Bioelectric sensors, Neural interfaces - Microelectrode technology, Neural interfaces - Biomaterials
Abstract: The lack of high-fidelity nerve interfaces continues to be a challenge for the field, and even more so in the context of chronic, awake experiments. The dorsal root ganglia (DRG) is an especially useful target to electrophysiologists because large amplitude afferent activity related to autonomic and somatic systems are easily recorded there while all efferent (or motor) axons route through the ventral root. However, a major challenge to the DRG in larger animals like cat, pig, dog, and human is the difficulty to penetrate the epineurium and, unlike dura mater, its removal is not a practical option. This talk will focus on our development and testing of a novel approach to interface the DRG, specifically designed to map peripheral nerve afferents. Here we present our success at developing ultra-fine, flexible arrays (5-µm thick and 70-µm wide) having 64 channels per shank. These were successfully inserted into DRG using a novel, ultra-fine diamond shuttle that was later removed. The array recorded a variety of afferents in a cat model and we are developing a backpack for chronic use.

Keywords: Image analysis and classification - Machine learning / Deep learning approaches, Image feature extraction, Deformable image registration
Abstract: Deformable image registration (DIR) is a key technic for dose accumulation in radiotherapy. With our recent developed DIR tools for accurate dose summation, a machine learning and DIR based toxicity modeling is needed to correlate rectal toxicity and delivered dose for prostate cancer patients.

Keywords: Image registration, segmentation, compression and visualization - Machine learning / Deep learning approaches, Image segmentation
Abstract: High quality segmentation is desirable in radiation therapy (RT) treatment planning because RT treatment quality heavily relies on the accuracy of the target and organs delineation. In current RT clinical practice, manual contouring is still the standard, even though it is time consuming and prone to intra- and inter-observer variation. Various methods have been and are being developed for seeking accurate and efficient automatic contouring. In this mini-symposium, we will focus on reviewing the application of artificial intelligence (AI)-based auto-segmentation methods in RT application.

Keywords: PET and SPECT Imaging applications, Functional image analysis
Abstract: Modern radiotherapy is delivered with enough precision that the sources of error are driven by defining the correct tissue volume to treat and balancing tumor control with normal tissue toxicity. PET-CT imaging is routinely used for staging cancer patients, but has also been used to identify tumor tissue for lung, head and neck, and pelvic cancers. The next generation of PET-CT imaging applications will improve definition and characterization of tumor tissues for dose escalation and monitor the response of both tumor and normal tissues to improve patient outcomes with limited toxicity.

Keywords: Ultrasound imaging - Interventional, Image segmentation, Deformable image registration
Abstract: The main challenge for ultrasound (US)-guided CT-based prostate brachytherapy is to accurately define tumor and prostate in CT images. We propose to incorporate MRI-defined tumor into US images to guide HDR catheter placement, and also integrate MRI-defined tumor and US-defined prostate volume into CT-based treatment planning to guide accurate dose delivery. This US-guided HDR brachytherapy will make the radiation delivery more accurate and boost more focal, improve prostate delineation, enable accurate dose planning and delivery, and potentially enhance prostate HDR treatment outcome.

Keywords: Dual-energy X-ray imaging, CT imaging applications
Abstract: The advantages of dual energy computed tomography (DECT) include additional tissue composition information, improved soft tissue contrast and reduced beam hardening artifact, when compared to single energy CT. Because of its potential advantages, DECT has found increasing attention in radiation oncology departments and been investigated for various radiotherapy applications such as proton range estimation, brachytherapy dose calculation and functional lung-sparing treatment planning. This presentation will review the application of DECT in proton therapy, namely proton range estimation.

Keywords: Magnetic resonance imaging - Interventional MRI, Magnetic resonance imaging - Pulse sequence, Magnetic resonance imaging - Parallel MRI
Abstract: Brachytherapy is an important cancer treatment option that offers localized, high-dose radiation with less damage to surrounding tissues. For dosimetry verification, X-ray CT is commonly used for imaging and localizing the brachytherapy seeds. However, CT scans cause additional ionizing radiations and often contain metal artifacts. MRI has the benefits of showing pathological information. However, its application in this area is severely limited because conventional MRI shows the brachytherapy as dark voids or with metal artifacts, i.e., negative contrast. Our recent work demonstrated that a new method to provide positive contrast MR images of the seeds which provides superior visualization and localization than the conventional MRI. In this paper, some recent progress on accelerated imaging and reconstruction are reported.

Keywords: Neural stimulation, Neurological disorders - Stroke, Neurological disorders - Mechanisms
Abstract: The discovery of neural stem cells in the adult brain has spurred much interest in their potential use in regenerative medicine strategies. I will discuss the current state of knowledge pertaining to their fundamental biology as well as their potential role in neural repair. I will discuss the work we are doing to exploit and enhance the inherent properties of these cells to promote tissue regeneration and functional recovery, focusing on the application of electric fields for cell migration. I will discuss our current work exploring the role of electrical stimulation on the migration of endogenous neural precursor cells in vitro and in vivo and how environmental cues can influence the rapid and directed migration of neural precursor cells.

Keywords: Motor neuroprostheses - Robotics, Neurorehabilitation, Human performance - Sensory-motor
Abstract: Neural effects of robot-assisted stepping on excitabilities of spinal reflexes and corticospinal tract will be shown in this presentation. In a series of experiments with a robotic gait trainer, we revealed that peripheral afferent inputs have inhibitory effects on the spinal stretch reflexes and facilitatory effects on the corticospinal tract. Those data will be discussed in relation to robot-assisted gait training for patients with neurological disorders.

Keywords: Motor neuroprostheses, Neuromuscular systems - Central mechanisms, Neurorehabilitation
Abstract: This lecture will present neurophysiological effects of electrical stimulation of muscles, nerves, and the spinal cord on the central nervous system. Specifically, short- and long-term excitabilities of the spinal, corticospinal, and cortical mechanisms after electrical stimulation will be discussed in the context of neuroplasticity and rehabilitation.

Keywords: Neural stimulation - Deep brain, Brain physiology and modeling, Neurological disorders
Abstract: An understanding of the underlying physiological mechanisms of deep brain stimulation is essential in guiding its advancement. This lecture will discuss how DBS interacts with the brain on an electrophysiological level and how physiological information can be utilized to optimize this technology.

Keywords: Motor neuroprostheses - Prostheses, Neurological disorders, Motor learning, neural control, and neuromuscular systems
Abstract: In this lecture the functional electrical stimulation therapy (FEST) for restoring upper limb function in stroke and spinal cord injury individuals with severe upper limb paralysis will be presented. In the presentation we will cover: (i) the basic concepts of the FEST technology, (ii) patient selection process, (iii) therapy delivery process, (iv) anticipated outcomes, and (v) basic physiology principles behind the therapy.

Keywords: Magnetic resonance imaging - Cardiac imaging, Fetal and Pediatric Imaging, Cardiac imaging and image analysis
Abstract: The assumption of blood as a Newtonian fluid is common in computational and experimental studies of cardiovascular biofluid mechanics, however, may not be valid in settings where low-shear flow is more prevalent. The Fontan procedure for univentricular heart defects creates an abnormal circulation where pulmonary arterial blood flow is non-pulsatile and low-shear. We explored the effect of non-Newtonian behavior on flow in synthetic models of the Fontan circulation using 4D flow magnetic resonance imaging.

Keywords: Magnetic resonance imaging - Cardiac imaging, Optical imaging, Cardiac imaging and image analysis
Abstract: The first cardiovascular intrinsic frequencies (ω1) were extracted from carotid waveforms obtained noninvasively using applanation tonometry. Various measures of left ventricle (LV) contractility were also measured from cardiac MRI data. The results indicate that noninvasively measured ω1 can be used a surrogate of LV contractility.

Keywords: Magnetic resonance imaging - Cardiac imaging, Magnetic resonance imaging - MR angiographic imaging, Magnetic resonance imaging - Pulse sequence
Abstract: Cardiac MRI is increasingly recognized as an important imaging modality for assessing cardiac anatomy, function and myocardial tissue composition. In this work, we describe a 3D dynamic cardiac MRI technique that is capable of producing high quality cardiac images in the presence of steady state intravascular distribution of ferumoxytol as an MRI contrast agent.

Keywords: Magnetic resonance imaging - Pulse sequence, Magnetic resonance imaging - Cardiac imaging
Abstract: 4D Flow has demonstrated clinical application in a variety of regions within the body, and evidence of its utility continues to grow. We will provide an overview of current methods and techniques and point to opportunities and challenges.

Keywords: Wearable robotic prosthetics, Joint biomechanics, Biologically inspired robotics and micro-biorobotics - Modeling
Abstract: Wearable robotic systems are developed to closely replicate the mechanics of the intact human body. For lower extremity applications, kinetics and kinematics are used to guide design; however, this approach is incomplete, and does not account for the limb’s mechanical impedance. Several studies have investigated the mechanical impedance of the human ankle joint, but these data have not yet been unified into a single control law, which is the purpose of this talk.

Keywords: Biologically inspired robotics and micro-biorobotics - Biologically inspired locomotion, Biologically inspired robotics and micro-biorobotics - Modeling, Neural control of movement and robotics applications
Abstract: The goal of this paper in the context of the Minisymposium on how to teach robots to move: lessons from biological motor control are an update on the neural control of bipedal walking in relation to bioinspired robots. This update will also be used to illustrate the concept of biomimetism and the dual relation between Biology and Engineering. First, we will review the biological concepts that have inspired robotic walking, such as the pattern generators. At a lower control level there is the Central Pattern Generator (CPG). Several biped robots use a symmetrical CPG. New evidence suggests that CPG behaves asymmetrically with its flexor half linked more tightly to the rhythm generator. Afterwards, we will look at an aspect of bipedal robots that is being used to understand human gait: a stability criterion. At a higher control level, the stability of bipedal gait is an important problem for robots and biological systems. While it is not easy to determine how biological biped systems guarantee stability, robot solutions can be useful to propose new hypothesis for biology.

Keywords: Motor learning, neural control, and neuromuscular systems, Neuromuscular systems - Computational modeling, Neuromuscular systems - Peripheral mechanisms
Abstract: Temporal expansion and nonlinear parameter varying methods are two promising approaches to the identification of the time-varying properties of joint stiffness during functional activities. This paper describes each method, demonstrates their application, discusses their strengths and weaknesses, and raises some open areas of research.

Keywords: Motor learning, neural control, and neuromuscular systems, Neuromuscular systems - Peripheral mechanisms, Neuromuscular systems - Locomotion
Abstract: The mechanical properties of our joints enable and constrain our interaction with the physical world. One mechanical property often characterized is the impedance, as it characterizes the dynamic relationship between imposed changes in position and the resultant torque. Recent studies have begun characterizing joint impedance during movement with various distinct methods. It remains unclear which method is ideal for estimating joint impedance during the time-varying conditions that occur during movement. In this mini-symposium, I will present the use of the multi-segment algorithm for estimating time-varying joint impedances. The multi-segment algorithm is ideally suited for the task of estimating time-varying impedances, as it is non-parametric, makes no assumption on the nature of the time-varying behavior, and can estimate both rapidly and slowly varying systems.

Keywords: Neuromuscular systems - Postural and balance, Human performance - Engineering, Human performance - Modelling and prediction
Abstract: Identification of joint impedance during movement requires the use of time-varying (TV) system identification techniques. Over the past years several new TV algorithms have been developed. Here, we adopted three novel techniques to identify TV joint impedance. All three techniques successfully estimated TV behavior from a single experimental trial.

Keywords: Motor learning, neural control, and neuromuscular systems
Abstract: Humans can make complex movements and are able to tune the mechanical properties of their joints. System identification is an emerging tool to assess joint dynamics in a quantitative way in both healthy and pathological conditions. During movement and task transitions, humans adapt their joint impedance. To investigate human movement and understand the origin of this adaptation, time-varying system identification techniques are required. The last years several novel time-varying algorithms have been developed with each their specific application. Here, we give an overview of the requirements for time-varying techniques to investigate human motion control and propose to develop a benchmark. In the ideal case a time-varying algorithm is able to track fast changes, at least as fast as humans can adapt, can be applied on a single trial of data to investigate trial-by-trial variability, and requires little to no prior information.

Keywords: Microfluidic applications, BioMEMS/NEMS - Tissue engineering and biomaterials
Abstract: Here we report development and fabrication of a microvessel containing Liver-on-a-Chip system to mimic liver function and use it for drug screening and disease modeling.

Keywords: Scaffolds in tissue engineering - Self-assembled, Scaffolds in tissue engineering - Patterned 3D, Biomaterial-cell interactions - Engineered vascular tissue
Abstract: Our knowledge on human brain development in normal and disease condition is limited due to limited accessibility. The possibility of creating human cerebral organoids from human induced pluripotent stem cell (hIPSC) presents a novel and exciting opportunity to investigate the mechanisms underlying normal brain development and severe neuropsychiatric disorders such as autism, epilepsy, schizophrenia, depression, and intellectual disability etc. Here we report our approaches and progress in disease model selection, hiPSC based disease modeling, and a microphysiological system (MPS) with cerebral organoids (CO) and micro-vessels to study the disease mechanisms underlying neuropsychiatric diseases and its potential applications for drug screening.

Keywords: Micro and Nano formulation - Microparticles/Microspheres, Micro and Nano formulation - Microencapsulation, Pharmaceutical engineering
Abstract: The achievement of desirable pharmacokinetic parameters from particulate drug delivery systems are dependent on the physical characteristics of the systems namely, particle dimension, loading of therapeutic agent, encapsulation efficiency, in vitro release kinetics. This study aimed to evaluate the main and interaction effects of the formulation variables on those physical characteristics and also to optimize the best combination of the variables to formulate small size particles with high encapsulation efficiency. The results showed that all the process variables (amount of polycaprolactone and stirring speed) except the amount of surfactant contributed significantly to the parameters previously mentioned. The best optimized formulation was experimentally validated for the closeness to the theoretical estimates.

Keywords: Pharmaceutical engineering
Abstract: Laparoscopic surgery has the advantage of the minimally invasive for patients. However, the surgery is technically difficult for surgeon because high dexterity is required for suturing in the narrow patient’s body. This paper presents a sealing method to locate the adhesive plaster at the incision instead of suturing. The objective is to optimize the plaster material and structure. We made the plaster with the thermally cross-linked gelatin film in a spiral fold because thermally cross-linked gelatin film has the high biocompatibility and tackiness, and a spiral fold has great storage efficiency. In 3 experiments, we measured expansion rate, expansion tension, peeling force, and sealing pressure in a variety of gelatin volume and concentration, and the films diameter. From these experimental results, we optimized the films using response surface method. As a result, the plaster is optimal at gelatin volume 10 mL, gelatin concentration 4 wt %, and films diameters 75 mm. We concluded that the optimized spiral folded adhesive plaster is sufficient in terms of the expansion, tackiness, and sealing properties.

Keywords: Drug delivery systems and carriers - Peptide and protein drug delivery, Clinical pharmacology - Pharmacokinetics/Pharmacodynamics, Pharmaceutical engineering
Abstract: The in-trans delivery of protein therapeutics across the blood-brain barrier by K16ApoE peptide carrier has been demonstrated to improve the neurological symptoms and increase the life-span of late-infantile neuronal ceroid lipofuscinosis (LINCL) mice. However, acute toxicity of K16ApoE was observed in LINCL mice resulting in a narrow therapeutic index, limiting the potential of translating the K16ApoE into a viable drug delivery system. This study aims to unravel the toxic mechanism of action. We hypothesized that the toxic response towards the peptide was induced by inhibition of acetylcholinesterase (AChE) activity at neuro-muscular junction. Here, results from the dose-response study suggested that AChE activity was inhibited by K16ApoE at either low or high doses but not at the mid-dose where a significant increase in AChE activity was observed. Meanwhile, molecular docking simulations showed that the N-terminus of K16ApoE is capable of binding to the active site gorge of AChE. In addition to a favorable spatial orientation, this docking pose also revealed strong surface charge interactions which may account for the observed inhibitory effect. While statistical analysis of the dose response and survival ratio suggested that AChE is not the primary mechanism of action for the acute toxicity of K16ApoE, both biochemical evidence and structural analysis have assigned indirect but critical roles for AChE in the overall toxicity mechanism of this peptide carrier.

Keywords: Clinical pharmacology - Pharmacokinetics/Pharmacodynamics, Clinical pharmacology - Computer simulation/modeling
Abstract: Some of commercial continuous glucose monitoring (CGM) devices, i.e., minimally-invasive sensors able to measure almost continuously glucose concentration in the subcutaneous tissue, recently received the regulatory approval to be used for making therapeutic decisions in diabetes management. A fundamental requirement for its safe and effective use is represented by the accuracy of CGM measurements. However, despite recent advances in sensors accuracy and reliability, CGM still suffers from inaccuracy problems in presence of pharmacologic interferences, e.g., the common orally administered acetaminophen (APAP), which artificially raises CGM glucose readings for several hours. A model of the artifact induced by APAP on CGM measurements would be useful to design algorithms to compensate such a distortion. The aim of this work is to exploit the data published by previous literature studies to design a model of oral APAP pharmacokinetics and its effect on glucose concentration measured by CGM sensors. Specifically, the developed model was identified on average data of both plasma APAP concentration and the APAP effect on CGM profiles after an oral administration of 1000 mg of APAP. The APAP effect on CGM readings was estimated from the difference observed, in the same study, between the glucose profile measured by a Dexcom G4 Platinum sensor and the plasma glucose concentration. The model was validated by comparing the simulated effect of mealtime APAP administration in CGM measurements of 100 virtual subjects generated by the UVA/Padova Type 1 Diabetes (T1D) Simulator vs. the effect observed in a clinical study by Maahs et al. (Diabetes Care, 2015) in 40 T1D subjects taking APAP at breakfast. Results suggest that the proposed model is able to reliably describe the mean APAP effect on CGM measurements.

Keywords: Data mining and processing - Pattern recognition, Time-frequency and time-scale analysis - Wavelets, Signal pattern classification
Abstract: Periodic limb movement disorder (PLMD) is a sleep disorder characterized by repetitive limb movements (LM) during night. The gold standard for LM detection consists of visual analysis of tibialis left (TIBL) and right (TIBR) electromyographic (EMG) signals. Such analysis is subjective and time-consuming. We here propose a semi-supervised and data-driven approach for LM detection during sleep that was trained and tested on 27 healthy controls (C) and 36 PLMD patients. After preprocessing of the EMG signals, discrete wavelet transform (Daubechies 4 mother wavelet and down to 4th decomposition level) was applied. EMG was reconstructed for each set of detail coefficients, thus obtaining four signals (D1-D4). The pre-processed EMG and D1-D4 signals were divided in 3-s mini-epochs of which traditional EMG features were calculated. Based on the assumption of lack of movements in healthy controls during rapid eye movement (REM) sleep, we used the features during REM of a subgroup of C to build a non-parametric probabilistic model defining the resting EMG distribution. This model was then used to classify the remaining mini-epochs as either resting EMG or LM. The percentages of 3-s mini-epochs with LMs were calculated for each subject and used to distinguish the remaining C and PLMD with a support vector machine and 5-fold cross validation scheme. Results showed that C can be distinguished by PLMD with accuracy higher than 82% in the preprocessed EMG and D1-D3 signals.

Keywords: Nonlinear dynamic analysis - Biomedical signals, Physiological systems modeling - Signal processing in physiological systems, Data mining and processing - Pattern recognition
Abstract: Pediatric sleep apnea-hypopnea syndrome (SAHS) is a highly prevalent respiratory disorder that may impose many negative effects on the health and development of children. Due to the drawbacks of overnight polysomnography (PSG), the gold standard diagnosis technique, automated analysis of nocturnal oximetry has emerged as a simplified alternative. In order to improve diagnosis ability of oximetry, we propose to evaluate the usefulness of AdaBoost, a classification boosting algorithm, in the context of pediatric SAHS. A database composed of 981 SpO2 recordings from pediatric subjects was used. For this purpose, a signal processing approach divided into two main stages was conducted: (i) feature extraction, where 3% oxygen desaturation index (ODI3), spectral, and nonlinear features were computed from the oximetry signal, and (ii) AdaBoost classification, where an AdaBoost.M2 model was trained with these features to determine the severity of pediatric SAHS according to the apnea-hypopnea index (AHI): AHI<1 events per hour (e/h), 1≤AHI<5 e/h, and AHI≥5 e/h. Our AdaBoost.M2 model achieved a Cohen’s kappa of 0.474 in an independent test set in the 3-class classification task. In addition, high accuracies were obtained when using the AHI cutoffs for diagnosis of mild (AHI=1 e/h) and moderate-to-severe (AHI=5 e/h) SAHS: 80.9% and 82.9%, respectively. These results outperformed ODI3 as well as state-of-the-art studies. Therefore, AdaBoost could help to enhance the diagnostic ability of the oximetry signal to assess pediatric SAHS severity.

Keywords: Neural networks and support vector machines in biosignal processing and classification, Signal pattern classification, Data mining and processing in biosignals
Abstract: Current sleep medicine relies on the analysis of polysomnographic measurements, comprising amongst others electroencephalogram (EEG), electromyogram (EMG), and electrooculogram (EOG) signals. This analysis currently requires supervision of a trained expert. Convolutional neural networks (CNN) provide an interesting framework to automated classification of sleep epochs based on raw EEG, EOG and EMG waveforms. In this study, we apply CNN approaches from the literature to four databases from pathological and physiological subjects. The best performing model resulted in Cohen's Kappa of k = 0.75 on healthy subjects and k = 0.64 on patients suffering from a variety of sleep disorders. Further, we show the advantages of additional sensor data (i.e. EOG and EMG). Deep learning approaches require a lot of data which is scarce for less prevalent diseases. For this, we propose a transfer learning procedure by pretraining a model on large public data and fine-tune this on each subject from a smaller dataset. This procedure is demonstrated using a private REM Behaviour Disorder database, improving sleep classification by 24.4%.

Keywords: Signal pattern classification, Neural networks and support vector machines in biosignal processing and classification, Data mining and processing - Pattern recognition
Abstract: This study aims at assessing the bispectral analysis of blood oxygen saturation (SpO2) from nocturnal oximetry to help in pediatric sleep apnea-hypopnea syndrome (SAHS) diagnosis. Recent studies have found excessive redundancy in the SAHS-related information usually extracted from SpO2, while proposing only two features as a reduced set to be used. On the other hand, it has been suggested that SpO2 bispectral analysis is able to provide complementary information to common anthropometric, spectral, and clinical variables. We address these novel findings to assess whether bispectrum provides new non-redundant information to help in SAHS diagnosis. Thus, we use 981 pediatric SpO2 recordings to extract both the reduced set of features recently proposed as well as 9 bispectral features. Then, a feature selection method based on the fast correlation-based filter and bootstrapping is used to assess redundancy among all the features. Finally, the non-redundant ones are used to train a Bayesian multi-layer perceptron neural network (BY-MLP) that estimate the apnea-hypopnea index (AHI), which is the diagnostic reference variable. Bispectral phase entropy was found complementary to the two previously recommended features and a BY-MLP model trained with the three of them reached high agreement with actual AHI (intra-class correlation coefficient = 0.889). Estimated AHI also showed high diagnostic ability, reaching 82.1%, 81.9%, and 90.3% accuracies and 0.814, 0.880, and 0.922 area under the receiver-operating characteristics curve for three common AHI thresholds: 1 e/h, 5 e/h, and 10 e/h, respectively. These results suggest that the information extracted from the bispectrum of SpO2 can improve the diagnostic performance of the oximetry test.

Keywords: Physiological systems modeling - Signal processing in physiological systems, Time-frequency and time-scale analysis - Time-frequency analysis
Abstract: Obstructive Sleep Apnea (OSA) is the most common form of sleep-disordered breathing in children. The gold standard to screen for OSA, polysomnography (PSG), requires an overnight stay in the hospital and is resource intensive. The Phone Oximeter is a non-invasive smartphone-based tool to record pulse oximetry. This portable device is able to measure patients over multiple nights while at home, causing less sleep disturbance than PSG and is able to measure night to night variability in sleep. This study analyzed the Screen My Sleep children (SMS) dataset, in which 74 children were monitored over multiple nights with the Phone Oximeter, including one night simultaneously with PSG in the hospital and two nights at home. In this study, we aim to investigate the night to night variability and assess the accuracy of the oxygen desaturation index (ODI) screening for children with significant OSA. In order to assess the performance of the ODI calculation in children, we implemented different ODIs at different desaturation levels and time durations. The variability was studied using a one-way ANOVA, and ODI’s performance screening for OSA using the area under the ROC curve (AUC). The implemented ODIs provide similar OSA screening results, using different apnea/hypopnea index (AHI) thresholds, as the ODI recommended for adults by the American academy of sleep medicine (AASM). The ODI provides an AUC of around 0.77, 0.76, 0.94 and 0.97 classifying children with an AHI>1, AHI>5 AHI>10 and AHI>15, respectively. The SMS dataset shows no significant night to night variability between the two nights at home. However, when comparing with the night at the hospital, both nights at home show a decrease in the lowest SpO2 value as well as overall SpO2 signal quality percentage. This study shows that there is variability in SpO2 signal between at-home versus in hospital settings.

Keywords: Neural networks and support vector machines in biosignal processing and classification, Data mining and processing in biosignals
Abstract: Traditional manual scoring of the entire sleep for diagnosis of sleep disorders is highly time-consuming and dependent to experts’ experience. Thus, automatic methods based on electrooculography (EOG) analysis have been increasingly attracted attentions to lower the cost of scoring. Such computer-aided diagnosis of sleep disorders are usually based on the 6 scores, wake (W), sleep status (S1--S4) and REM by labelling every 30-second long EOG records. This paper presents an automatic scoring method of sleep stages by using the recent advancements in deep learning. We also suggest an interactive scoring scheme to enable the doctors of practitioners to give feedback by correcting errors and improve the accuracy of scoring as well as diagnosis of sleep disorders.

Keywords: Vascular mechanics and hemodynamics - Arterial pressure in cardiovascular disease, Vascular mechanics and hemodynamics - Pulse wave velocity, Cardiovascular and respiratory signal processing - Pulse transit time
Abstract: We developed a smartphone-based device for cuff-less and calibration-free measurement of systolic and diastolic blood pressure (BP). We prospectively tested the device against standard automatic cuff BP measurements in human subjects. The smartphone-based device yielded bias errors within 5.4 mmHg in magnitude and precision errors within 9.0 mmHg over a 40-50 mmHg range of BP.

Keywords: Cardiovascular and respiratory signal processing - Blood pressure measurement
Abstract: Applanation tonometry has been used in the past for non-cuff based blood pressure measurements. However, the use of this technique is limited in the ambulatory setting, as well as with non-trained personnel. We present results from a preliminary feasibility study on using a two-dimensional force sensor array for blood pressure measurements by untrained subjects.

Keywords: Cardiovascular and respiratory signal processing - Pulse transit time, Cardiovascular and respiratory signal processing - Blood pressure measurement, Vascular mechanics and hemodynamics - Pulse wave velocity
Abstract: Pulse transit time (PTT) through central arteries was measured by a developed patch-type wearable device and blood pressure was estimated based on measured PTT. Validation was carried out during exercise and the PTT based estimation showed good agreement with reference blood pressure.

Keywords: Cardiovascular and respiratory signal processing - Blood pressure measurement, Vascular mechanics and hemodynamics - Arterial pressure in cardiovascular disease
Abstract: State-of-the-art blood pressure measurement devices suffer from various shortcoming for the athlete; they are either occlusive, invasive, inaccurate, hard-to-use, or operator dependent. Our objective is to create a blood pressure measurement device which provides a way to easily acquire frequent measurements. The proposed approach is based on force-measured ultrasound images of a subject’s carotid artery. A tissue-deformation model-fitting optimization estimates systolic and diastolic blood pressure. First tests on a volunteer suggest that the approach may have application for rapidly estimating both systolic and diastolic blood pressures.

Keywords: Wearable body sensor networks and telemetric systems, Modeling and analysis, Novel methods
Abstract: Gait of Parkinson's disease patients delivers important information to support diagnosis and therapy decisions. To maximize the usability and interpretability of this information, data analytics and machine learning tools are needed that facilitate gait parameter calculation and clinical scale development. In this presentation, general considerations regarding the application of wearable sensor-based assessment systems as well as our own work towards the development of an automatic score in Parkinson's Disease (PD) are discussed.

Keywords: Novel methods
Abstract: We sought to employ a machine learning approach for the automatic recognition of health problems that manifest themselves in distinctive behaviors during varying locomotor tasks. In this paper, we provide an illustration of our analysis to full body movement in response to a destabilizing gait perturbation task of trip. Movement patterns were captured using a motion-capture system, yielding time series of bilateral lower limb, pelvic, and trunk joint angles. Initially, an off-line supervised One versus All machine learning algorithm was used to classify the pathological state. The algorithm classifies the user’s movements into healthy and stroke with hemiparesis. Each event was categorized into three subtasks; pre-trip, trip, and post-trip. By breaking down movements into subtasks, one can review whether the sub-classifications for each subject agree as a verification construct. The experimental results achieved classification accuracies near 100% depending on the subtask. Our future goal is to explore if the high classification accuracy seen herein can serve as a predictive model that can subsequently be used to automatically recognize health states in subjects performing a completely novel task.

Keywords: Wearable body sensor networks and telemetric systems, Modeling and analysis, Physiological monitoring - Modeling and analysis
Abstract: The world is experiencing an unprecedented aging process, which leads to the increased occurrence of gait disorders due to age-related neurological disease and thus the increased demand for gait physical therapy. However, current gait rehabilitative therapy is expensive, subjective and inefficient and thus the aim of this work is to design and develop an integrative sensing, learning and analytics framework to automate the gait physical therapy process with objective detection, evaluation and recommendation capabilities.

Keywords: Image registration, segmentation, compression and visualization - Machine learning / Deep learning approaches
Abstract: We present a machine learning method for extracting gait parameters, such as cadence, step length, peak knee flexion, and Gait Deviation Index, from a single video. The method leverages a set of annotated videos and an algorithm for two-dimensional pose estimation (OpenPose). Our predictions using a convolution neural network model achieved correlation ⍴=0.74 with GDI computed from optical motion capture.

Keywords: Neural stimulation, Neural interfaces - Implantable systems
Abstract: The pancreas is a highly innervated organ which receives neuromodulatory signals from both autonomic and sensory inputs. Efferent signaling by the vagus nerve enters the pancreas and synapses onto intrapancreatic ganglion. Post-ganglionic fibers directly innervate islets to control hormone secreting cells including β-cells which secrete insulin. Type 1 diabetes (T1D) is a chronic condition which results in the inability of the body to regulate glucose homeostasis after major, but not complete β-cell death. In order to elucidate the possible role of therapeutic parasympathetic modulation in T1D, electrical stimulation of the vagus nerve was delivered to diabetic and control rats.

Keywords: Neural stimulation, Neural interfaces - Implantable systems, Neurological disorders - Treatment methodologies
Abstract: A challenge in bioelectric medicine is avoiding off-target effects of neural stimulation. One reason for off-target effects is the fact that peripheral and autonomic nerve stimulation excites both afferent and efferent nerves, with impulses traveling in both directions from the site of stimulation. Here we review recent work pairing a high-frequency blocking electrode with a stimulating electrode to force nerve activation to only propagate in one direction, enabling a directional and virtual vagotomy. We discuss recent results using this method to enhance the anti-inflammatory effects of vagal nerve stimulation.

Keywords: Neural stimulation, Sensory neuroprostheses
Abstract: Mammalian blood glucose concentrations (i.e. glycemia) are maintained within well-defined biological limits despite considerable fluctuations in the rate at which glucose is obtained from food and utilized by tissues. Failure to achieve glycemia (e.g., normal blood glucose levels) can result in significant physiological complications such as diabetes and hypertension, clinically referred to Type 2 Diabetes (T2D). Glycemia is regulated by both hormonal and neural control mechanisms, with a majority of investigations focused on hormonal regulation. Recent investigations suggest that chronically increased sympathetic input to the visceral organs contributes to the development and progression of T2D. We investigated the role of sympathetic input to the visceral organs by inhibiting activity using Kilohertz Electrical Stimulation (KES) nerve block.

Keywords: Neural interfaces - Implantable systems, Motor neuroprostheses - Neuromuscular stimulation, Neural interfaces - Neuromorphic engineering
Abstract: Ventilation is an autonomic function that relies on a finely tuned biological control system that dynamically adapts to maintain homeostasis in the body. Neurotrauma or neuropathologies such as spinal cord injury, stroke, central hypoventilation syndrome, Pompe disease, and central sleep apnea can impair ventilatory control. Restoration of adequate ventilation often relies on mechanical ventilation. However, mechanical ventilators have considerable drawbacks that can sometimes further compromise ventilation such as alveolar damage and diaphragm atrophy. Targeting the neural innervation for ventilatory control through bioelectronic interfaces and biologically-inspired controllers offers an alternate approach to restore ventilation. We will discuss the design and use of such neuromorphic controller for ventilatory pacing.

Keywords: Neural signals - Coding, Neural interfaces - Microelectrode technology
Abstract: Over the last two decades we have seen significant advances in neural decoding and stimulation technologies which have opened a door to future diagnostic and treatment options. Neural interfaces including highly flexible cuff microelectrode arrays for recording and stimulation have allowed increasingly detailed study of neural circuits and mechanisms and more advanced neuromodulation approaches for manipulating organ function and even treating injury. Decoding and rerouting signals around injured portions of the nervous system are now possible. This has allowed a paralyzed man to move his hand again and perform movements that are useful in daily life. These and other developments are leading to the rapid expansion of a new field called bioelectronic medicine which is bringing a new arsenal for battling disease and injury in patients around the world.

Keywords: Sensory neuroprostheses - Visual, Neural interfaces - Implantable systems
Abstract: We present a mobile object recognition framework developed for the Argus II Retinal Prosthesis System. An explore mode enables users to quickly identify the presence of unkown objects within an environment, while a search mode applies auditory cueing to help users pin-point specific objects of interest with ease. Using an NVIDIA Jetson TX2 for embedded computing, our system achieves an average object recognition frame rate of 5 frames per second using the YOLOv2 network and provides a runtime of about 3 hours.

Keywords: Brain-computer/machine interface, Sensory neuroprostheses - Visual, Neural stimulation
Abstract: We present an energy-optimum multichannel neuroprosthetic device using three custom built chips, namely, 4-channel stimuli-generator (SG), 16-channel microelectrode driver (MED) and rectifier, which are designed in 0.13μm CMOS and 0.8μm CMOS/DMOS technologies respectively, along with some commercial chips. The SG is able to generate rectangular, half-sine, plateau-sine and other types of stimulation-efficient current pulses with the maximum value of 220μA per channel. The MED delivers up to 20V at its output channels, when supplied with ±13V. The prototype has been assembled with a platinum coated pyramidal 3D microelectrode array. The measured microelectrode average impedance is 70kΩ at 1kHz from in-vitro test results with 0.9% Phosphate buffer saline

Keywords: Sensory neuroprostheses - Visual
Abstract: Visual scanning by sighted individuals is done using eye and head movements. In contrast, scanning using the Argus II is solely done by head movement, since eye movements can introduce localization errors. It was demonstrated that a scanning mode utilizing eye movements increases the performances of a visual prosthesis. In the lecture we will present and discuss technical challenges, specifically how to calibrate an eye tracker for blind users. An integrated eye tracker in a visual prosthesis will measure gaze position in real-time that will be used to shift the region of interest (ROI) that is sent to the implant within the wide field of view (FOV) of the scene camera. Users will be able to use combined eye-head scanning: shifting the camera by moving their head and shifting the ROI within the FOV by eye movement. But, traditional eye tracker calibration methods requiring looking at points in space and cannot be used with blind people. We demonstrated that by correlating the pupil location at the onset of the stimulation with the head-centered percept location we can calibrate and align the eye tracker for Argus II users. Our experimental results show that integrating a calibrated eye tracker reduces the amount of head motion and improves visual stability in Argus II users.

Keywords: Sensory neuroprostheses - Visual, Human performance - Activities of daily living
Abstract: Retinal prostheses for the blind can restore the perception of light for individuals with severe blindness. The ability of retinal implant patients to recognize objects is limited due to the low resolution of current implants. Computer vision has the potential to perform object recognition and path planning, and augment overall patient performance.

Keywords: Brain physiology and modeling - Neural dynamics and computation, Brain physiology and modeling - Neuron modeling and simulation
Abstract: Activity in cortical networks is notable for its temporal coherence, the dominant frequency bands of which are relevant markers of brain state and health. Understanding the mechanistic bases of this activity is a critical step in beginning to understand how to treat many neurological illnesses. We have approached this problem using a wide variety of methods, including computational modeling; dynamical systems theory; cellular electrophysiology; real-time, low-latency feedback systems; and calcium imaging.

Keywords: Neural signal processing, Neural stimulation, Neural signals - Coding
Abstract: One clear test as to whether we truly understand the neural code is whether we can “tap” into the activity of the neurons and make clear predictions about what is going on in the outside world – to “read the neural code”. Conversely, as a community, we are finally in a position to advance methodologies for controlling neuronal circuits in detail: To “write the neural code”. With the rapid acceleration of tools, we now see the emergence of real time reading and writing of the neural code through closed-loop control strategies that enable us to have a bi-directional conversation with the brain.

Keywords: Brain physiology and modeling - Neural dynamics and computation
Abstract: Spikes generated in the presence of epileptogenic agent 4_AP can propagate in the hippocampus non-synaptically. Experiments also rule out gap junctions, diffusion and axonal transmission to explain the propagation. Ephaptic coupling is tested below using computer simulation as a possible mechanism of propagation in a flat array of pyramidal cells. Propagation could be observed at similar speeds to those recorded experimentally. In addition the model shows that low amplitude fields can be sufficient to excite other cells and generate self-propagating non-synaptic waves

Keywords: Drug delivery routes - Transdermal drug delivery, Drug delivery systems and carriers - Peptide and protein drug delivery, Pharmaceutical engineering
Abstract: Dissolving microneedles have a high potential to replace hypodermic needles and topical formulations. The current technologies however are limited in delivery efficiency due to various physical and chemical factors. By employing optimized dissolving microneedle fabrication techniques, it is possible to maintain biological activity of encapsulated compounds at high levels. Here we introduce few methods that we have developed to increase the delivery efficiency of dissolving microneedles.

Keywords: Pharmaceutical engineering, Micro and Nano formulation - Microparticles/Microspheres, Drug delivery routes - Transdermal drug delivery
Abstract: Chronic drug delivery to deep brain region was achieved by MEMS neural probe integrated with a microfluidic channel. Also, the integrated wireless module and micropump enabled wireless drug delivery system.

Keywords: IT in Pharmaceutical R&D - Artificial Intelligence in pharmaceutical & clinical research, IT in Pharmaceutical R&D - Real-World Data and applications
Abstract: Clinical trials are critical to determining the safety and efficacy of new potential drugs and treatments for patients, playing a vital role in improving health care. According to a report by Industry Standard Research, patient recruitment and enrollment inefficiencies are the primary causes of delays on a new drug development, costing the sponsor 600K to 8M in potential revenue [1]. The rapid increase in the number of clinical trials and more complex protocols have contributed to this challenge. Patient eligibility screening is still conducted manually by research personnel in a very laborious and inefficient manner. Health record data have been proved to be the useful source for clinical trial recruitment [2]. With the advanced technology of machine learning including natural language processing (NLP) and information extraction (IE), which automatically extract designated clinical data from text, we sought to develop an automated eligibility assessment algorithm for clinical trials using health record. This algorithm has the potential to significantly increase the efficiency and performance standard of patient eligibility screening, contributing to better and more economic conduct of clinical trials. The gained efficiency will accelerate new drug development to treat patients with diseases.

Keywords: Computer model-based assessments for regulatory submissions, Computer modeling for treatment planning
Abstract: Recently, the Task Group 103 of the International Commission on Radiological Protection (ICRP) has completed the development of new mesh-type reference computational phantoms (MRCPs). Compared to the current voxel-type reference computational phantoms (VRCPs), the MRCPs provide many advantages including deformability, which makes it possible to create phantoms in different statures and postures for personalized dose calculation. In this respect, in the present study, the MRCPs were deformed in different statures and postures. The deformed phantoms were then used to calculate dose coefficients for industrial radiography sources near the body, which can be used as a first estimator of organ doses of individuals who are accidentally exposed to an industrial radiography source.

Keywords: Computer model-based assessments for regulatory submissions, Computer modeling for treatment planning
Abstract: Over the past 20 years, use of computational human phantoms within existing Monte Carlo radiation transport code required those phantoms to be in a voxelized format. Recently, codes such as MCNP, PHITS, and GEANT4 now allow the transport in human computational phantoms represented by polygon mesh structures. While both phantoms provide a high degree of anatomic realism compared to first-generation stylized (or mathematical phantoms), mesh-type phantoms are now considered state-of-the-art and permits re-sculpting of individual organs, body circumferences, body size, body shape, and extremity articulation – all features not readily available to voxel-based phantoms. Over the past two decades a tremendous number of voxel-based phantoms have been developed from CT or MR data, and thus there is a need for conversion of existing models to mesh-type formats to allow these additional benefits. The objective of this study is to develop an efficient and accurate methodology to convert voxel-based phantoms to mesh-based phantoms. For this conversion, a boundary detection algorithm is implemented in conjunction with polygon detection to form high-quality meshed data suitable for radiation transport simulation and finite element analysis. This conversion can result in a reduction of required simulation time as well as allowing current voxel data to be used in modern CAD software.

Keywords: Computer model-based assessments for regulatory submissions, Computer modeling for treatment planning
Abstract: Recently, Computational human models with fine-resolution have been frequently used for various research fields. In this study, we present a technique to generate personalized computational human models and human growth models developed by a volume morphing.

Keywords: Clinical engineering, Computer modeling for treatment planning
Abstract: Abstract— Bone fracture healing is a four-phase process: inflammatory response, soft callus formation, hard callus development, and remodeling. Since bone healing is a very complex process, where healing outcome relies on various biological and mechanical factors, its mechanisms have not been completely understood. On the other hand, accurate prediction and pattern recognition of healing status at different phases of healing can be very helpful in clinical applications and design of treatment strategies. Therefore, employment of machine learning techniques and neural networks can be very useful in prediction and pattern recognition of the healing process, as there is no need for full knowledge of this complex process. In this study, we present four neural networks for four phases of healing to process bone healing status at each its phase.

Keywords: Clinical laboratory, assay and pathology technologies
Abstract: Abstract — Osteoporotic fractures pose a significant challenge in developing countries due to an aging population coupled with limited availability of resources. The goal of the proposed work is to use plain X-rays, a universally accessible imaging modality, to identify image based indices and patient specific parameters that predict hip fracture. We will employ a software program called X-ray based Fracture Prediction Tool (XFx) to establish an accurate, cost-effective, and accessible platform to predict fracture risk.

Keywords: Computer model-based assessments for regulatory submissions, Computer modeling for treatment planning, Health technology - Verification and validation
Abstract: In many current and emerging biomedical applications, precise numerical models that take into account the inhomogeneous lossy nature of the human body are required to enable the ability of researchers and designers to optimize solutions in this complex medium. To this end, a respiratory sequence capturing the biomechanics involved in normal breathing has been created and integrated into the CAD based NELLY computational phantom. Construction of the model and its applications are discussed.