Filtering property of myelinated internode can change neural information representability and might trigger a compensatory action during demyelination.

In this paper, for the first time, we showed that an Internode Segment (INS) of a myelinated axon acts as a lowpass filter, and its filter characteristics depend on the number of myelin turns. Consequently, we showed how the representability of a neural signal could be altered with myelin loss in pathological conditions involving demyelinating diseases. Contrary to the traditionally held viewpoint that myelin geometry of an INS is optimised for maximising Conduction Velocity (CV) of Action Potential (AP), our theory provides an alternative viewpoint that myelin geometry of an INS is optimised for maximizing representability of the stimuli a fibre is meant to carry. Subsequently, we show that this new viewpoint could explain hitherto unexplained experimentally observed phenomena such as, shortening of INS length during demyelination and remyelination, and non-uniform distribution of INS in the central nervous system fibres and associated changes in diameter of nodes of ranvier along an axon. Finally, our theory indicates that a compensatory action could take place during demyelination up to a certain number of loss of myelin turns to preserve the neural signal representability by simultaneous linear scaling of the length of an INS and the inner radius of the fibre.

Personalized subcutaneous implantable cardioverter-defibrillator sensing vectors generated by mathematical rotation increase device eligibility whilst preserving device performance.

AIMS: Approximately 5.7% of potential subcutaneous implantable cardioverter-defibrillator (S-ICD) recipients are ineligible by virtue of their vector morphology, with higher rates of ineligibility observed in some at-risk groups. Mathematical vector rotation is a novel technique that can generate a personalized sensing vector, one with maximal R:T ratio, using electrocardiogram (ECG) signal recorded from the present S-ICD location. METHODS AND RESULTS: A cohort of S-ICD ineligible patients were identified through ECG screening of ICD patients with no ventricular pacing requirement and their personalized vectors were generated using ECG signal from a Holter monitor. Subcutaneous ICD eligibility in this cohort was then recalculated. In a separate cohort, episodes of arrhythmia were recorded in patients undergoing arrhythmia induction, and arrhythmia detection in standard S-ICD vectors was compared to rotated vectors using an S-ICD simulator. Ninety-two participants (mean age 64.9 ± 2.7 years) underwent screening and 5.4% were found to be S-ICD ineligible. Personalized vector generation increased the R:T ratio in these vectors from 2.21 to 7.21 (4.54-9.88, P < 0.001) increasing the cohort eligibility from 94.6% to 100%. Rotated S-ICD vectors also showed high ventricular fibrillation (VF) detection sensitivity (97.8%), low time to VF detection (6.1 s), and excellent tachycardia discrimination (sensitivity 96%, specificity 88%), with no significant differences between rotated and standard vectors. CONCLUSION: In S-ICD ineligible patients, mathematical vector rotation can generate a personalized vector that is associated with a significant increase in R:T ratio, resulting in universal device eligibility in our cohort. Ventricular fibrillation detection efficacy, time to VF detection, and tachycardia discrimination were not affected by vector rotation.

Prediction and classification of ventricular arrhythmia based on phase-space reconstruction and fuzzy c-means clustering.

BACKGROUND AND OBJECTIVE: Prediction and classification of Ventricular Arrhythmias (VA) may allow clinicians sufficient time to intervene for stopping its escalation to Sudden Cardiac Death (SCD). This paper proposes a novel method for predicting VA and classifying its type, in particular, the fatal VA even before the event occurs. METHODS: A statistical index J based on the combination of phase-space reconstruction (PSR) and box counting has been used to predict VA. The fuzzy c-means (FCM) clustering technique is applied for the classification of impending VA. RESULTS: 32 healthy and 32 arrhythmic subjects from two open databases - PTB Diagnostic database (PTBDB) and CU Ventricular Tachyarrhythmia (CUDB) database respectively; were used to validate our proposed method. Our method showed average prediction time of approximately 5 min (4.97 min) for impending VA in the tested dataset while classifying four types of VA (VA without ventricular premature beats (VPBs), ventricular fibrillation (VF), ventricular tachycardia (VT), and VT followed by VF) with an average 4 min (approximately) before the VA onset, i.e., after 1 min of the prediction time point with average accuracy of 98.4%, a sensitivity of 97.5% and specificity of 99.1%. CONCLUSIONS: The results obtained can be used in clinical practice after rigorous clinical trial to advance technologies such as implantable cardioverter defibrillator (ICD) that can help to preempt the occurrence of fatal ventricular arrhythmia - a main cause of SCD.

A new algorithm to reduce T-wave over-sensing based on phase space reconstruction in S-ICD system.

BACKGROUND AND OBJECTIVE: The subcutaneous implantable cardioverter defibrillator (S-ICD) reduces mortality in individuals at high risk of sudden arrhythmic death, by rapid defibrillation of life-threatening arrhythmia. Unfortunately, S-ICD recipients are also at risk of inappropriate shock therapies, which themselves are associated with increased rates of mortality and morbidity. The commonest cause of inappropriate shock therapies is T wave oversensing (TWOS), where T waves are incorrectly counted as R waves leading to an overestimation of heart rate. It is important to develop a method to reduce TWOS and improve the accuracy of R-peak detection in S-ICD system. METHODS: This paper introduces a novel algorithm to reduce TWOS based on phase space reconstruction (PSR); a common method used to analyse the chaotic characteristics of non-linear signals. RESULTS: The algorithm was evaluated against 34 records from University Hospital Southampton (UHS) and all 48 records from the MIT-BIH arrhythmia database. In the UHS analysis we demonstrated a sensitivity of 99.88%, a positive predictive value of 99.99% and an accuracy of 99.88% with reductions in TWOS episodes (from 166 to 0). Whilst in the MIT-BIH analysis we demonstrated a sensitivity of 99.87%, a positive predictive value of 99.99% and an accuracy of 99.91% for R wave detection. The average processing time for 1 min ECG signals from all records is 2.9 s. CONCLUSIONS: Our algorithm is sensitive for R-wave detection and can effectively reduce the TWOS with low computational complexity, and it would therefore have the potential to reduce inappropriate shock therapies in S-ICD recipients, which would significantly reduce shock related morbidity and mortality, and undoubtedly improving patient's quality of life.

Classification of Autism Spectrum Disorder From EEG-Based Functional Brain Connectivity Analysis.

Autism is a psychiatric condition that is typically diagnosed with behavioral assessment methods. Recent years have seen a rise in the number of children with autism. Since this could have serious health and socioeconomic consequences, it is imperative to investigate how to develop strategies for an early diagnosis that might pave the way to an adequate intervention. In this study, the phase-based functional brain connectivity derived from electroencephalogram (EEG) in a machine learning framework was used to classify the children with autism and typical children in an experimentally obtained data set of 12 autism spectrum disorder (ASD) and 12 typical children. Specifically, the functional brain connectivity networks have quantitatively been characterized by graph-theoretic parameters computed from three proposed approaches based on a standard phase-locking value, which were used as the features in a machine learning environment. Our study was successfully classified between two groups with approximately 95.8% accuracy, 100% sensitivity, and 92% specificity through the trial-averaged phase-locking value (PLV) approach and cubic support vector machine (SVM). This work has also shown that significant changes in functional brain connectivity in ASD children have been revealed at theta band using the aggregated graph-theoretic features. Therefore, the findings from this study offer insight into the potential use of functional brain connectivity as a tool for classifying ASD children.

Linear and nonlinear analysis of intrinsic mode function after facial stimuli presentation in children with autism spectrum disorder.

In this work, a method for classifying Autism Spectrum Disorders (ASD) from typically developing (TD) children is presented using the linear and nonlinear Event-Related Potential (ERP) analysis of the Electro-encephalogram (EEG) signals. The signals were acquired during the presentation of three types of face expression stimuli -happy, fearful and neutral faces. EEGs are first decomposed using the Multivariate Empirical Mode Decomposition (MEMD) method to extract its Intrinsic Mode Functions (IMFs), which provide information about the underlying activities of ERP components. The nonlinear sample entropy (SampEn) features, as well as the standard linear measurements utilizing maximum (Max.), minimum (Min), and standard deviation (Std.), are then extracted from each set of IMFs. These features are then evaluated by the statistical analysis tests and used to construct the input vectors for the Discriminant analysis (DA), Support vector machine (SVM), and k-Nearest Neighbors (kNN) classifiers. Experimental results show that the proposed features can differentiate the ASD and TD children using the happy stimulus dataset with high classification performance for all classifiers that reached 100% accuracy. This result suggests a general deficit in recognizing the positive expression in ASD children. Additionally, we found that the SampEn measurements computed from the alpha and theta bands and the linear features extracted from the delta band can be considered biomarkers for disturbances in Emotional Facial Expression (EFE) processing in ASD children.

Cognitive Outcome Prediction in Infants With Neonatal Hypoxic-Ischemic Encephalopathy Based on Functional Connectivity and Complexity of the Electroencephalography Signal.

Impaired neurodevelopmental outcome, in particular cognitive impairment, after neonatal hypoxic-ischemic encephalopathy is a major concern for parents, clinicians, and society. This study aims to investigate the potential benefits of using advanced quantitative electroencephalography analysis (qEEG) for early prediction of cognitive outcomes, assessed here at 2 years of age. EEG data were recorded within the first week after birth from a cohort of twenty infants with neonatal hypoxic-ischemic encephalopathy (HIE). A proposed regression framework was based on two different sets of features, namely graph-theoretical features derived from the weighted phase-lag index (WPLI) and entropies metrics represented by sample entropy (SampEn), permutation entropy (PEn), and spectral entropy (SpEn). Both sets of features were calculated within the noise-assisted multivariate empirical mode decomposition (NA-MEMD) domain. Correlation analysis showed a significant association in the delta band between the proposed features, graph attributes (radius, transitivity, global efficiency, and characteristic path length) and entropy features (Pen and SpEn) from the neonatal EEG data and the cognitive development at age two years. These features were used to train and test the tree ensemble (boosted and bagged) regression models. The highest prediction performance was reached to 14.27 root mean square error (RMSE), 12.07 mean absolute error (MAE), and 0.45 R-squared using the entropy features with a boosted tree regression model. Thus, the results demonstrate that the proposed qEEG features show the state of brain function at an early stage; hence, they could serve as predictive biomarkers of later cognitive impairment, which could facilitate identifying those who might benefit from early targeted intervention.

Hardware architecture for real-time EEG-based functional brain connectivity parameter extraction.

Objective. Design a novel architecture for real-time quantitative characterization of functional brain connectivity (FC) networks derived from wearable electroencephalogram (EEG).Approach. We performed an algorithm to architecture mapping for the calculation of phase lag index to form the functional connectivity networks and the extraction of a set of graph-theoretic parameters to quantitatively characterize these networks. This mapping was optimized using approximations in the mathematical definitions of the algorithms which reduce its computational complexity and produce a more hardware amenable implementation.Main results. The architecture was developed for a 19-channel EEG system. The system can calculate all the functional connectivity parameters in a total time of 131 µs, utilizes 71% of the total logic resources in the FPGA, and shows 51.84 mW dynamic power consumption at 22.16 MHz operation frequency when implemented in a Stratix IV EP4SGX230K FPGA. Our analysis also showed that the system occupies an area equivalent to approximately 937 K 2-input NAND gates, with an estimated power consumption of 39.3 mW at 0.9 V supply using a 90 nm CMOS application specific integrated circuit technology.Significance. The proposed architecture can calculate the FC and extract the graph-theoretic parameters in real-time with low power consumption. This characteristic makes the architecture ideal for applications such as a wearable closed-loop neurofeedback systems, where constant monitoring of the brain activity and fast processing of EEG is necessary to control the appropriate feedback.

An automated toolchain for quantitative characterisation of structural connectome from MRI based on non-anatomical cortical parcellation.

Brain connectivity analysis is a new multidisciplinary approach in neuroscience for determining neurological disorders from brain imaging data. But, there is no end-to-end toolchain that processes raw MRI data and extracts brain connectivity network metrics. Again, the existing method of cortical parcellation from MRI data is mainly based on fixed Brodmann atlas; which does not support neonate's brain or adult's brain with neuroplasticity anomalies. In this work, we design an end-to-end toolchain that processes raw MRI data and generates network metrics for brain connectivity analysis using non-anatomical equal-area parcellation. We process the structural and diffusion MRI data to generate the parcellated and segmented image, extract white matter tracks and build structural connectome and then interface it with Brain Connectivity Toolbox to extract graph theory measures.Clinical relevance An automated tool for end-to-end processing of MRI data to brain connectivity pattern extraction and its quantitative characterisation for diagnosing brain disorder.

An Automatic R and T Peak Detection Method Based on the Combination of Hierarchical Clustering and Discrete Wavelet Transform.

The detection and delineation of QRS-complexes and T-waves in Electrocardiogram (ECG) is an important task because these features are associated with the cardiac abnormalities including ventricular arrhythmias that may lead to sudden cardiac death. In this paper, we propose a novel method for the R-peak and the T-peak detection using hierarchical clustering and Discrete Wavelet Transform (DWT) from the ECG signal. In the first step, a template of the single ECG beat is identified. Secondly, all R-peaks are detected by using hierarchical clustering. Then, each corresponding T-wave boundary is delineated based on the template morphology. Finally, the determination of T wave peaks is achieved based on the Modulus-Maxima Analysis (MMA) of the DWT coefficients. We evaluated the algorithm by using all records from the MIT-BIH arrhythmia database and QT database. The R-peak detector achieved a sensitivity of 99.89%, a positive predictivity of 99.97% and 99.83% accuracy over the validation MIT-BIH database. In addition, it shows a sensitivity of 100%, a positive predictivity of 99.83% in manually annotated QT database. It also shows 99.92% sensitivity and 99.96% positive predictivity over the automatic annotated QT database. In terms of the T-peak detection, our algorithm is verified with 99.91% sensitivity and 99.38% positive predictivity in manually annotated QT database.

Phase Space Reconstruction Based CVD Classifier Using Localized Features.

This paper proposes a generalized Phase Space Reconstruction (PSR) based Cardiovascular Diseases (CVD) classification methodology by exploiting the localized features of the ECG. The proposed methodology first extracts the ECG localized features including PR interval, QRS complex, and QT interval from the continuous ECG waveform using features extraction logic, then the PSR technique is applied to get the phase portraits of all the localized features. Based on the cleanliness and contour of the phase portraits CVD classification will be done. This is first of its kind approach where the localized features of ECG are being taken into considerations unlike the state-of-art approaches, where the entire ECG beats have been considered. The proposed methodology is generic and can be extended to most of the CVD cases. It is verified on the PTBDB and IAFDB databases by taking the CVD including Atrial Fibrillation, Myocardial Infarction, Bundle Branch Block, Cardiomyopathy, Dysrhythmia, and Hypertrophy. The methodology has been tested on 65 patients' data for the classification of abnormalities in PR interval, QRS complex, and QT interval. Based on the obtained statistical results, to detect the abnormality in PR interval, QRS complex and QT interval the Coefficient Variation (CV) should be greater than or equal to 0.1012, 0.083, 0.082 respectively with individual accuracy levels of 95.3%, 96.9%, and 98.5% respectively. To justify the clinical significance of the proposed methodology, the Confidence Interval (CI), the p-value using ANOVA have been computed. The p-value obtained is less than 0.05, and greater F-statistic values reveal the robust classification of CVD using localized features.

Rehab-Net: Deep Learning Framework for Arm Movement Classification Using Wearable Sensors for Stroke Rehabilitation.

In this paper, we present a deep learning framework "Rehab-Net" for effectively classifying three upper limb movements of the human arm, involving extension, flexion, and rotation of the forearm, which, over the time, could provide a measure of rehabilitation progress. The proposed framework, Rehab-Net is formulated with a personalized, light weight and low-complex, customized convolutional neural network (CNN) model, using two-layers of CNN, interleaved with pooling layers, followed by a fully connected layer that classifies the three movements from tri-axial acceleration input data collected from the wrist. The proposed Rehab-Net framework was validated on sensor data collected in two situations: 1) semi-naturalistic environment involving an archetypal activity of "making-tea" with four stroke survivors and 2) natural environment, where ten stroke survivors were free to perform any desired arm movement for the duration of 120 min. We achieved an overall accuracy of 97.89% on semi-naturalistic data and 88.87% on naturalistic data which exceeded state-of-the-art learning algorithms namely, linear discriminant analysis, support vector machines, and k-means clustering with an average accuracy of 48.89%, 44.14%, and 27.64%. Subsequently, a computational complexity analysis of the proposed model has been discussed with an eye toward hardware implementation. The clinical significance of this study is to accurately monitor the clinical progress of the rehabilitated subjects under the ambulatory settings.

A phase lag index hardware calculation for real-time electroencephalography studies.

Among the different techniques used for the analysis of electroencephalograms, the phase lag index has become an important method for the calculation of the functional brain connectivity. Currently, this method is implemented offline due to its high computational complexity restricting it from real-time applications that would require an online neurofeedback. In this paper, we propose a new architecture to calculate the phase lag index of electroencephalograms in real-time. As a proof of concept, a 32 bit and 16-channel system running at 188.32 MHz was synthesized on a Stratix IV GX FPGA. The system was tested and the simulations demonstrated that the system could perform the calculation of the Phase lag index at least 66 times faster than the MATLAB software with a mean square error of less than 5.72×10-6.

Low-complexity hardware design methodology for reliable and automated removal of ocular and muscular artifact from EEG.

BACKGROUND AND OBJECTIVE: EEG is a non-invasive tool for neuro-developmental disorder diagnosis and treatment. However, EEG signal is mixed with other biological signals including Ocular and Muscular artifacts making it difficult to extract the diagnostic features. Therefore, the contaminated EEG channels are often discarded by the medical practitioners which may result in less accurate diagnosis. Many existing methods require reference electrodes, which will create discomfort to the patient/children and cause hindrance to the diagnosis of the neuro-developmental disorder and Brain Computer Interface in the pervasive environment. Therefore, it would be ideal if these artifacts can be removed real time on the hardware platform in an automated fashion and then the denoised EEG can be used for online diagnosis in a pervasive personalized healthcare environment without the need of any reference electrode. METHODS: In this paper we propose a reliable, robust and automated methodology to solve the aforementioned problem. The proposed methodology is based on the Haar function based Wavelet decompositions with simple threshold based wavelet domain denoising and artifacts removal schemes. Subsequently hardware implementation results are also presented. 100 EEG data from Physionet, Klinik für Epileptologie, Universität Bonn, Germany, Caltech EEG databases and 7 EEG data from 3 subjects from University of Southampton, UK have been studied and nine exhaustive case studies comprising of real and simulated data have been formulated and tested. The proposed methodology is prototyped and validated using FPGA platform. RESULTS: Like existing literature, the performance of the proposed methodology is also measured in terms of correlation, regression and R-square statistics and the respective values lie above 80%, 79% and 65% with the gain in hardware complexity of 64.28% and improvement in hardware delay of 53.58% compared to state-of-the art approaches. Hardware design based on the proposed methodology consumes 75 micro-Watt power. CONCLUSIONS: The automated methodology proposed in this paper, unlike the state of the art methods, can remove blink and muscular artifacts real time without the need of any extra electrode. Its reliability and robustness is also established after exhaustive simulation study and analysis on both simulated and real data. We believe the proposed methodology would be useful in next generation personalized pervasive healthcare for Brain Computer Interface and neuro-developmental disorder diagnosis and treatment.

CNN based approach for activity recognition using a wrist-worn accelerometer.

In recent years, significant advancements have taken place in human activity recognition using various machine learning approaches. However, feature engineering have dominated conventional methods involving the difficult process of optimal feature selection. This problem has been mitigated by using a novel methodology based on deep learning framework which automatically extracts the useful features and reduces the computational cost. As a proof of concept, we have attempted to design a generalized model for recognition of three fundamental movements of the human forearm performed in daily life where data is collected from four different subjects using a single wrist worn accelerometer sensor. The validation of the proposed model is done with different pre-processing and noisy data condition which is evaluated using three possible methods. The results show that our proposed methodology achieves an average recognition rate of 99.8% as opposed to conventional methods based on K-means clustering, linear discriminant analysis and support vector machine.

GOLIAH (Gaming Open Library for Intervention in Autism at Home): a 6-month single blind matched controlled exploratory study.

BACKGROUND: To meet the required hours of intensive intervention for treating children with autism spectrum disorder (ASD), we developed an automated serious gaming platform (11 games) to deliver intervention at home (GOLIAH) by mapping the imitation and joint attention (JA) subset of age-adapted stimuli from the Early Start Denver Model (ESDM) intervention. Here, we report the results of a 6-month matched controlled exploratory study. METHODS: From two specialized clinics, we included 14 children (age range 5-8 years) with ASD and 10 controls matched for gender, age, sites, and treatment as usual (TAU). Participants from the experimental group received in addition to TAU four 30-min sessions with GOLIAH per week at home and one at hospital for 6 months. Statistics were performed using Linear Mixed Models. RESULTS: Children and parents participated in 40% of the planned sessions. They were able to use the 11 games, and participants trained with GOLIAH improved time to perform the task in most JA games and imitation scores in most imitation games. GOLIAH intervention did not affect Parental Stress Index scores. At end-point, we found in both groups a significant improvement for Autism Diagnostic Observation Schedule scores, Vineland socialization score, Parental Stress Index total score, and Child Behavior Checklist internalizing, externalizing and total problems. However, we found no significant change for by time × group interaction. CONCLUSIONS: Despite the lack of superiority of TAU + GOLIAH versus TAU, the results are interesting both in terms of changes by using the gaming platform and lack of parental stress increase. A large randomized controlled trial with younger participants (who are the core target of ESDM model) is now discussed. This should be facilitated by computing GOLIAH for a web platform. Trial registration Clinicaltrials.gov NCT02560415.

GOLIAH: A Gaming Platform for Home-Based Intervention in Autism - Principles and Design.

Children with Autism need intensive intervention and this is challenging in terms of manpower, costs, and time. Advances in Information Communication Technology and computer gaming may help in this respect by creating a nomadically deployable closed-loop intervention system involving the child and active participation of parents and therapists. An automated serious gaming platform enabling intensive intervention in nomadic settings has been developed by mapping two pivotal skills in autism spectrum disorder: Imitation and Joint Attention (JA). Eleven games - seven Imitations and four JA - were derived from the Early Start Denver Model. The games involved application of visual and audio stimuli with multiple difficulty levels and a wide variety of tasks and actions pertaining to the Imitation and JA. The platform runs on mobile devices and allows the therapist to (1) characterize the child's initial difficulties/strengths, ensuring tailored and adapted intervention by choosing appropriate games and (2) investigate and track the temporal evolution of the child's progress through a set of automatically extracted quantitative performance metrics. The platform allows the therapist to change the game or its difficulty levels during the intervention depending on the child's progress. Performance of the platform was assessed in a 3-month open trial with 10 children with autism (Trial ID: NCT02560415, Clinicaltrials.gov). The children and the parents participated in 80% of the sessions both at home (77.5%) and at the hospital (90%). All children went through all the games but, given the diversity of the games and the heterogeneity of children profiles and abilities, for a given game the number of sessions dedicated to the game varied and could be tailored through automatic scoring. Parents (N = 10) highlighted enhancement in the child's concentration, flexibility, and self-esteem in 78, 89, and 44% of the cases, respectively, and 56% observed an enhanced parents-child relationship. This pilot study shows the feasibility of using the developed gaming platform for home-based intensive intervention. However, the overall capability of the platform in delivering intervention needs to be assessed in a bigger open trial.

Hybrid wavelet and EMD/ICA approach for artifact suppression in pervasive EEG.

BACKGROUND: Electroencephalogram (EEG) signals are often corrupted with unintended artifacts which need to be removed for extracting meaningful clinical information from them. Typically a priori knowledge of the nature of the artifacts is needed for such purpose. Artifact contamination of EEG is even more prominent for pervasive EEG systems where the subjects are free to move and thereby introducing a wide variety of motion-related artifacts. This makes hard to get a priori knowledge about their characteristics rendering conventional artifact removal techniques often ineffective. NEW METHOD: In this paper, we explore the performance of two hybrid artifact removal algorithms: Wavelet Packet Transform followed by Independent Component Analysis (WPTICA) and Wavelet Packet Transform followed by Empirical Mode Decomposition (WPTEMD) in pervasive EEG recording scenario, assuming existence of no a priori knowledge about the artifacts and compare their performance with two existing artifact removal algorithms. RESULTS: Artifact cleaning performance has been measured using Root Mean Square Error (RMSE) and Artifact to Signal Ratio (ASR)-an index similar to traditional Signal to Noise Ratio (SNR), and also by observing normalized power distribution topography over the scalp. COMPARISON WITH EXISTING METHOD(S): Comparison has been made first using semi-simulated signals and then with real experimentally acquired EEG data with commercially available 19-channel pervasive EEG system Enobio corrupted by eight types of artifact. CONCLUSIONS: Our explorations show that WPTEMD consistently gives best artifact cleaning performance not only in semi-simulated scenario but also in the case of real EEG data containing artifacts.

Detecting Elementary Arm Movements by Tracking Upper Limb Joint Angles With MARG Sensors.

This paper reports an algorithm for the detection of three elementary upper limb movements, i.e., reach and retrieve, bend the arm at the elbow and rotation of the arm about the long axis. We employ two MARG sensors, attached at the elbow and wrist, from which the kinematic properties (joint angles, position) of the upper arm and forearm are calculated through data fusion using a quaternion-based gradient-descent method and a two-link model of the upper limb. By studying the kinematic patterns of the three movements on a small dataset, we derive discriminative features that are indicative of each movement; these are then used to formulate the proposed detection algorithm. Our novel approach of employing the joint angles and position to discriminate the three fundamental movements was evaluated in a series of experiments with 22 volunteers who participated in the study: 18 healthy subjects and four stroke survivors. In a controlled experiment, each volunteer was instructed to perform each movement a number of times. This was complimented by a seminaturalistic experiment where the volunteers performed the same movements as subtasks of an activity that emulated the preparation of a cup of tea. In the stroke survivors group, the overall detection accuracy for all three movements was 93.75% and 83.00%, for the controlled and seminaturalistic experiment, respectively. The performance was higher in the healthy group where 96.85% of the tasks in the controlled experiment and 89.69% in the seminaturalistic were detected correctly. Finally, the detection ratio remains close ( ±6%) to the average value, for different task durations further attesting to the algorithms robustness.

Dynamical System Approach for Edge Detection Using Coupled FitzHugh-Nagumo Neurons.

The prospect of emulating the impressive computational capabilities of biological systems has led to considerable interest in the design of analog circuits that are potentially implementable in very large scale integration CMOS technology and are guided by biologically motivated models. For example, simple image processing tasks, such as the detection of edges in binary and grayscale images, have been performed by networks of FitzHugh-Nagumo-type neurons using the reaction-diffusion models. However, in these studies, the one-to-one mapping of image pixels to component neurons makes the size of the network a critical factor in any such implementation. In this paper, we develop a simplified version of the employed reaction-diffusion model in three steps. In the first step, we perform a detailed study to locate this threshold using continuous Lyapunov exponents from dynamical system theory. Furthermore, we render the diffusion in the system to be anisotropic, with the degree of anisotropy being set by the gradients of grayscale values in each image. The final step involves a simplification of the model that is achieved by eliminating the terms that couple the membrane potentials of adjacent neurons. We apply our technique to detect edges in data sets of artificially generated and real images, and we demonstrate that the performance is as good if not better than that of the previous methods without increasing the size of the network.