Department Publications

Neonatal jaundice, caused by elevated bilirubin levels, requires timely diagnosis to prevent severe complications. Existing deep learning approaches primarily focus on binary classification of jaundice from visual data, but do not provide quantitative bilirubin estimation. A binary decision alone is insufficient for clinical management, where treatment thresholds depend on precise bilirubin levels. This paper presents a deep-learning-based contactless framework for neonatal jaundice assessment using image-based analysis. The proposed lightweight multi-stage pipeline integrates eye-region obfuscation, skin segmentation, and deep-learning-based image classification and bilirubin estimation. The system first performs jaundice classification; upon positive detection, a regression model is invoked to estimate bilirubin. Furthermore, image-based neonatal screening raises privacy concerns, as facial images of …

Remote photoplethysmography (rPPG) enables contact free monitoring of vital signs and is especially valuable for neonates, since conventional methods often require sustained skin contact with adhesive probes that can irritate fragile skin and increase infection control burden. We present VideoPulse, a neonatal dataset and an end to end pipeline that estimates neonatal heart rate and peripheral capillary oxygen saturation (SpO2) from facial video. VideoPulse contains 157 recordings totaling 2.6 hours from 52 neonates with diverse face orientations. Our pipeline performs face alignment and artifact aware supervision using denoised pulse oximeter signals, then applies 3D CNN backbones for heart rate and SpO2 regression with label distribution smoothing and weighted regression for SpO2. Predictions are produced in 2 second windows. On the NBHR neonatal dataset, we obtain heart rate MAE 2.97 bpm using 2 second windows (2.80 bpm at 6 second windows) and SpO2 MAE 1.69 percent. Under cross dataset evaluation, the NBHR trained heart rate model attains 5.34 bpm MAE on VideoPulse, and fine tuning an NBHR pretrained SpO2 model on VideoPulse yields MAE 1.68 percent. These results indicate that short unaligned neonatal video segments can support accurate heart rate and SpO2 estimation, enabling low cost non invasive monitoring in neonatal intensive care.

Dry-contact ear-electroencephalography (Ear-EEG) enables wearable neural monitoring. However, motion induced electrode-skin impedance (ESI) mismatches between electrodes can severely degrade signal quality. To the best of our knowledge, this paper presents the first proof-of-concept dry-contact EarEEG system that uses a driven-right-leg (DRL) stimulus for continuous ESI mismatch monitoring, enabling online adaptive motion artifact cancellation. A 1 kHz sinusoidal stimulus is injected through the DRL electrode. The resulting response to the injected carrier is separated from the EEG using bandpass filtering and demodulation, and then used to extract the ESI mismatch information as the reference input for a normalized least-mean-square adaptive filter followed by a Hampel filtering stage. To evaluate artifact suppression and preservation of neural activity, alpha-band EEG activity was analyzed involving four healthy participants performing head nodding, electrode tapping, and jaw clenching. The system achieved artifact power reductions of 6.5, 12.6, and 9.0 dB (77.6%, 92.8%, and 86.4%, respectively) while alpha-band modulation remained clearly observable after processing. This demonstrates the feasibility of DRL-stimulusbased ESI mismatch monitoring for motion artifact cancellation in wearable dry-contact Ear-EEG.

Meditation has been associated with a range of cognitive and physiological benefits. However, the underlying neural mechanisms of different meditation practices are not yet fully understood. This study investigates whether electroencephalogram (EEG) signals can be used to characterize and distinguish two commonly practiced meditation techniques, namely Samatha and Vipassana, which involve distinct cognitive processes, with Samatha emphasizing sustained concentration and Vipassana emphasizing mindful observation and insight. By extracting features such as band power, coherence and wavelet entropy from EEG signals recorded during pre-meditation resting, Samatha, and Vipassana states, we provide an assessment of how spectral power, signal complexity, and functional connectivity differentiate these practices. Preliminary results from experienced meditators indicate condition-dependent EEG differences, with the most pronounced effects observed in the delta frequency band. These results demonstrate the utility of EEG-based metrics for the objective characterization of meditation and advance our comprehension of the neural processes underlying different contemplative practices.

Cardiovascular diseases rank among the leading causes of mortality worldwide and the early identification of diseases is of paramount importance. This work focuses on developing a novel machine learning-based framework for early detection and classification of heart murmurs by analysing phonocardiogram signals. Our heart murmur detection and classification pipeline encompasses three classification settings. We first develop a set of methods based on transfer learning to determine the existence of heart murmurs and categorize them as present, absent, or unknown. If a murmur is present it will be classified as normal or abnormal based on its clinical outcome by using 1D convolution and audio spectrogram transformers. Finally, we use Wav2Vec encoder with raw audio data and AdaBoost abstain classifier for heart murmur quality identification. Heart murmurs are categorized based on their specific attributes …

Neonates are highly susceptible to seizures, often leading to short or long-term neurological impairments. However, clinical manifestations of neonatal seizures are subtle and often lead to misdiagnoses. This increases the risk of prolonged, untreated seizure activity and subsequent brain injury. Continuous video electroencephalogram (cEEG) monitoring is the gold standard for seizure detection. However, this is an expensive evaluation that requires expertise and time. In this study, we propose a convolutional neural network-based model for early prediction of neonatal seizures by distinguishing between interictal and preictal states of the EEG. Our model is patient-independent, enabling generalization across multiple subjects, and utilizes mel-frequency cepstral coefficient matrices extracted from multichannel EEG and electrocardiogram (ECG) signals as input features. Trained and validated on the Helsinki neonatal EEG dataset with 10-fold cross-validation, the proposed model achieved an average accuracy of 97.52%, sensitivity of 98.31%, specificity of 96.39%, and F1-score of 97.95%, enabling accurate seizure prediction up to 30 minutes before onset. The inclusion of ECG alongside EEG improved the F1-score by 1.42%, while the incorporation of an attention mechanism yielded an additional 0.5% improvement. To enhance transparency, we incorporated SHapley Additive exPlanations (SHAP) as an explainable artificial intelligence method to interpret the model and provided localization of seizure focus using scalp plots. The overall results demonstrate the model's potential for minimally supervised deployment in neonatal intensive care …

Objective Young children and infants, especially newborns, are highly susceptible to seizures, which, if undetected and untreated, can lead to severe long-term neurological consequences. Early detection typically requires continuous electroencephalography (cEEG) monitoring in hospital settings, involving costly equipment and highly trained specialists. This study presents a low-cost, active dry-contact electrode–based, adjustable electroencephalography (EEG) headset, combined with an explainable deep learning model for seizure detection from reduced-montage EEG, and a multimodal artifact removal algorithm to enhance signal quality. Methods EEG signals were acquired via active electrodes and processed through a custom-designed analog front end for filtering and digitization. The adjustable headset was fabricated using three-dimensional printing and laser cutting to accommodate varying head sizes …

Objective Neonates are highly susceptible to seizures, which can have severe long-term consequences if undetected and left untreated. Early detection is crucial and typically requires continuous electroencephalography (EEG) monitoring in a hospital setting, which is costly, inconvenient, and requires specialized experts for diagnosis. In this work, we propose a new low-cost active dry-contact electrode-based adjustable EEG headset, a new explainable deep learning model to detect neonatal seizures, and an advanced signal processing algorithm to remove artifacts to address the key aspects that lead to the underdiagnosis of neonatal seizures. Methods EEG signals are acquired through active electrodes and processed using a custom-designed analog front end (AFE) that filters and digitizes the captured EEG signals. The adjustable headset is designed using three-dimensional (3D) printing and laser cutting to …

Photoplethysmography (PPG) has emerged as a critical non-invasive technique for cardiovascular health monitoring through wearable devices, yet the physiological mechanisms underlying PPG signal formation are not completely understood. This affects the development of accurate diagnostic tools and personalized health monitoring systems. To overcome it, analysis of how cardiac output and other physiological factors are reflected in PPG, especially at the wrist, is required, and end-to-end PPG simulation methods have been developed for this purpose. However, these methods have limitations in either accuracy or computational efficiency. To address this gap, we present a comprehensive end-to-end framework for simulating wristbased PPG signals by integrating radial artery hemodynamics with optical light transport modeling through skin. In our approach, first, cardiac parameters are used to generate the …

Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the …

The neonatal period is the most vulnerable time for the development of seizures. Seizures in the immature brain lead to detrimental consequences, therefore require early diagnosis. The gold -standard for neonatal seizure detection currently relies on continuous video-EEG monitoring; which involves recording multi-channel electroencephalogram (EEG) alongside real-time video monitoring within a neonatal intensive care unit (NICU). However, video-EEG monitoring technology requires clinical expertise and is often limited to technologically advanced and resourceful settings. Cost-effective new techniques could help the medical fraternity make an accurate diagnosis and advocate treatment without delay. In this work, a novel explainable deep learning model to automate the neonatal seizure detection process with a reduced EEG montage is proposed, which employs convolutional nets, graph attention layers, and …

Cardiovascular diseases rank among the leading causes of mortality worldwide and the early identification of diseases is of paramount importance. This work focuses on developing a novel machine learning-based framework for early detection and classification of heart murmurs by analysing phonocardiogram signals. Our heart murmur detection and classification pipeline encompasses three classification settings. We first develop a set of methods based on transfer learning to determine the existence of heart murmurs and categorize them as present, absent, or unknown. If a murmur is present it will be classified as normal or abnormal based on its clinical outcome by using 1D convolution and audio spectrogram transformers. Finally, we use Wav2Vec encoder with raw audio data and AdaBoost abstain classifier for heart murmur quality identification. Heart murmurs are categorized based on their specific attributes …

Accurate and accessible cognitive load monitoring is essential for improving personalized learning and facilitating better mental health monitoring. Existing methods for measuring cognitive load can be intrusive, costly, and less effective for individuals with high melanin concentrations, emphasizing the need for an improved approach. This study presents a novel approach to monitoring the cognitive load using functional near-infrared spectroscopy (fNIRS) technology. A lightweight, cost-effective fNIRS device with a single optode was developed to record prefrontal cortex data, specially designed to suit the skin with high melanin concentration. Data from participants performing n-back tasks revealed significant differences between hemoglobin concentrations among 1-back, 2-back, and 3-back tasks via ANOVA analysis. To further explore cognitive load assessment, various machine learning models were …

Cortical ischaemic strokes result in cognitive deficits depending on the area of the affected brain. However, we have demonstrated that difficulties with attention and processing speed can occur even with small subcortical infarcts. Symptoms appear independent of lesion location, suggesting they arise from generalized disruption of cognitive networks. Longitudinal studies evaluating directional measures of functional connectivity in this population are lacking. We evaluated six patients with minor stroke exhibiting cognitive impairment 6–8 weeks post-infarct and four age-similar controls. Resting-state magnetoencephalography data were collected. Clinical and imaging evaluations of both groups were repeated 6- and 12 months later. Network Localized Granger Causality was used to determine differences in directional connectivity between groups and across visits, which were correlated with clinical …

Sleep plays a crucial role in the well-being of human lives. Traditional sleep studies using Polysomnography are associated with discomfort and often lower sleep quality caused by the acquisition setup. Previous works have focused on developing less obtrusive methods to conduct high-quality sleep studies, and ear- EEG is among popular alternatives. However, the performance of sleep staging based on ear-EEG is still inferior to scalp- EEG based sleep staging. In order to address the performance gap between scalp-EEG and ear- EEG based sleep staging, we propose a cross-modal knowledge distillation strategy 1 † https://github.com/Mithunjha/EarEEG_KnowledgeDistillation, which is a domain adaptation approach. We employ model architectures from the transformer and convolutional neural network families to demonstrate the model-agnostic nature of the method. Our experiments and analysis validate the …

Stroke patients with hemiparesis display decreased beta band (13–25 Hz) rolandic activity, correlating to impaired motor function. However, clinically, patients without significant weakness, with small lesions far from sensorimotor cortex, exhibit bilateral decreased motor dexterity and slowed reaction times. We investigate whether these minor stroke patients also display abnormal beta band activity. Magnetoencephalographic (MEG) data were collected from nine minor stroke patients (NIHSS < 4) without significant hemiparesis, at ~1 and ~6 months postinfarct, and eight age-similar controls. Rolandic relative beta power during matching tasks and resting state, and Beta Event Related (De)Synchronization (ERD/ERS) during button press responses were analyzed. Regardless of lesion location, patients had significantly reduced relative beta power and ERS compared to controls. Abnormalities persisted over visits, and were present in both ipsi- and contra-lesional hemispheres, consistent with bilateral impairments in motor dexterity and speed. Minor stroke patients without severe weakness display reduced rolandic beta band activity in both hemispheres, which may be linked to bilaterally impaired dexterity and processing speed, implicating global connectivity dysfunction affecting sensorimotor cortex independent of lesion location. Findings not only illustrate global network disruption after minor stroke, but suggest rolandic beta band activity may be a potential biomarker and treatment target, even for minor stroke patients with small lesions far from sensorimotor areas.

Human behavioral monitoring during sleep is essential for various medical applications. Majority of the contactless human pose estimation algorithms are based on RGB modality, causing ineffectiveness in in-bed pose estimation due to occlusions by blankets and varying illumination conditions. Long-wavelength infrared (LWIR) modality based pose estimation algorithms overcome the aforementioned challenges; however, ground truth pose generations by a human annotator under such conditions are not feasible. A feasible solution to address this issue is to transfer the knowledge learned from images with pose labels and no occlusions, and adapt it towards real world conditions (occlusions due to blankets). In this paper, we propose a novel learning strategy comprises of two-fold data augmentation to reduce the cross-domain discrepancy and knowledge distillation to learn the distribution of unlabeled images in …

Network slicing enables the creation of multiple logical independent networks on physical networking infrastructure. Network slice deployment in a Fifth Generation (5G) mobile network can be classified as vertical and horizontal slicing. The paper compares the performance of the two slicing methods through solving two convex optimization problems, considering several smart hospital scenarios that differ from each other based on their medical speciality. The results are used to draw insights on the most appropriate slicing approach for each setup.

Stroke patients with hemiparesis display decreased beta band (13–25Hz) rolandic activity, correlating to impaired motor function. However, clinically, patients without significant weakness, with small lesions far from sensorimotor cortex, exhibit bilateral decreased motor dexterity and slowed reaction times. We investigate whether these minor stroke patients also display abnormal beta band activity. Magnetoencephalographic (MEG) data were collected from nine minor stroke patients (NIHSS< 4) without significant hemiparesis, at~ 1 and~ 6 months postinfarct, and eight age-similar controls. Rolandic relative beta power during matching tasks and resting state, and Beta Event Related (De) Synchronization (ERD/ERS) during button press responses were analyzed. Regardless of lesion location, patients had significantly reduced relative beta power and ERS compared to controls. abnormalities persisted over visits, and were present in both ipsi-and contra-lesional hemispheres, consistent with bilateral impairments in motor dexterity and speed. Minor stroke patients without severe weakness display reduced rolandic beta band activity in both hemispheres, which may be linked to bilaterally impaired dexterity and processing speed, implicating global connectivity dysfunction affecting sensorimotor cortex independent of lesion location. Findings not only illustrate global network disruption after minor stroke, but suggest rolandic beta band activity may be a potential biomarker and treatment target, even for minor stroke patients with small lesions far from sensorimotor areas.

Diagnosing pre-existing heart diseases early in life is important as it helps prevent complications such as pulmonary hypertension, heart rhythm problems, blood clots, heart failure and sudden cardiac arrest. To identify such diseases, phonocardiogram (PCG) and electrocardiogram (ECG) waveforms convey important information. Therefore, effectively using these two modalities of data has the potential to improve the disease screening process. We evaluate this hypothesis on a subset of the PhysioNet Challenge 2016 Dataset which contains simultaneously acquired PCG and ECG recordings. Our novel dual-convolutional neural network based approach uses transfer learning to tackle the problem of having limited amounts of simultaneous PCG and ECG data that is publicly available, while having the potential to adapt to larger datasets. In addition, we introduce two main evaluation frameworks named record-wise …

Stroke patients with small central nervous system infarcts often demonstrate an acute dysexecutive syndrome characterized by difficulty with attention, concentration, and processing speed, independent of lesion size or location. We use magnetoencephalography (MEG) to show that disruption of network dynamics may be responsible. Nine patients with recent minor strokes and eight age-similar controls underwent cognitive screening using the Montreal cognitive assessment (MoCA) and MEG to evaluate differences in cerebral activation patterns. During MEG, subjects participated in a visual picture–word matching task. Task complexity was increased as testing progressed. Cluster-based permutation tests determined differences in activation patterns within the visual cortex, fusiform gyrus, and lateral temporal lobe. At visit 1, MoCA scores were significantly lower for patients than controls (median [interquartile range …

Network softwarization enables the novel concept of Local 5G Operator (L5GO) networks, for deploying localized 5G solutions to serve case and location specific communication needs. This paper proposes a L5GO network architecture for delay critical future telehealth services, considering two use cases on augmented reality assisted and robotic aided surgery. The paper compares the latency performance of the proposed L5GO architecture with a traditional legacy network and a network equipped with Multi-access Edge Computing (MEC). The results highlight the unique advantages of utilizing an L5GO to cater the communication needs of delay critical telehealth, compared to a traditional network.

Neural processing along the ascending auditory pathway is often associated with a progressive reduction in characteristic processing rates. For instance, the well-known frequency-following response (FFR) of the auditory midbrain, as measured with electroencephalography (EEG), is dominated by frequencies from~ 100 Hz to several hundred Hz, phase-locking to the stimulus waveform at those frequencies. In contrast, cortical responses, whether measured by EEG or magnetoencephalography (MEG), are typically characterized by frequencies of a few Hz to a few tens of Hz, time-locking to acoustic envelope features. In this study we investigated a crossover, cortically generated responses time-locked to continuous speech features at FFR-like rates. Using MEG, we analyzed high-frequency responses (70-300 Hz) to continuous speech using neural source-localized reverse correlation and its corresponding temporal response functions (TRFs). Continuous speech stimuli were presented to 40 subjects (17 younger, 23 older adults) with clinically normal hearing and their MEG responses were analyzed in the 70-300 Hz band. Consistent with the insensitivity of MEG to many subcortical structures, the spatiotemporal profile of these response components indicated a purely cortical origin with~ 40 ms peak latency and a right hemisphere bias. TRF analysis was performed using two separate aspects of the speech stimuli: a) the 70-300 Hz band of the speech waveform itself, and b) the 70-300 Hz temporal modulations in the high frequency envelope (300-4000 Hz) of the speech stimulus. The response was dominantly driven by the high frequency …

There are several systems that use one or several Kinect sensors for human gait analysis, particularly for diagnosis of patients. However, due to the limited depth sensing range of the Kinect—a sensor manufactured for video gaming—the depth measurement accuracy reduces with distance from the Kinect. In addition, self-occlusion of the subject limits the accuracy and utility of such systems. We overcome these limitations by using a two-Kinect gait analysis system and mechanically moving the Kinects in synchronization with the test subject and each other. This increases the practical measurement range of the Kinect-based system whilst maintaining the measurement accuracy. Results of the comparison of knee flexion, step length, and stride length with a software based method show that our moving Kinect system can accurately analyse these gait parameters.

This paper presents a novel low-complexity VLSI architecture for an EEG Feature Extraction Platform for wearable health monitoring systems. It integrates a processing core unit, preprocessing FIR filter block and an FFT accelerator. The platform supports 8 scalp EEG channels and it can be scaled up to 16 or 32 EEG channels. In this paper, we analyze several architectural optimizations for hardware implementation of features such as Power Spectral Density, Discrete Wavelet Transform, Autoregression etc. The proposed architecture is easily scalable to new feature extractors. An FFT accelerator is proposed to improve the flexibility of the platform. We provide a comparison with software models for each feature extractor and also validate the platform by implementing a seizure detection algorithm on CHB-MIT Scalp EEG database with an accuracy of over 95%.