Training size predictably improves machine learning-based epileptic seizure forecasting from wearables DOI Creative Commons
Mustafa Halimeh, Michele Jackson, Tobias Loddenkemper

et al.

Neuroscience Informatics, Journal Year: 2024, Volume and Issue: 5(1), P. 100184 - 100184

Published: Dec. 9, 2024

Language: Английский

Human Activity Recognition Algorithm with Physiological and Inertial Signals Fusion: Photoplethysmography, Electrodermal Activity, and Accelerometry DOI Creative Commons

Justin Gilmore,

Mona Nasseri

Sensors, Journal Year: 2024, Volume and Issue: 24(10), P. 3005 - 3005

Published: May 9, 2024

Inertial signals are the most widely used in human activity recognition (HAR) applications, and extensive research has been performed on developing HAR classifiers using accelerometer gyroscope data. This study aimed to investigate potential enhancement of models through fusion biological with inertial signals. The classification eight common low-, medium-, high-intensity activities was assessed machine learning (ML) algorithms, trained (ACC), blood volume pulse (BVP), electrodermal (EDA) data obtained from a wrist-worn sensor. Two types ML algorithms were employed: random forest (RF) features; pre-trained deep (DL) network (ResNet-18) spectrogram images. Evaluation conducted both individual more generalized groups, based similar intensity. Results indicated that RF outperformed corresponding DL at grouped levels. However, EDA BVP ACC improved classifier performance compared baseline model ACC-only best achieved by combination ACC, EDA, images, yielding F1-scores 69 87 for classifications, respectively. For additional signals, almost all classifications showed improvement (p-value < 0.05). In enhanced low- medium-intensity activities. Exploring two specific activities, ascending/descending stairs cycling, revealed significantly results combined BVP, images

Language: Английский

Citations

3
Autonomic biosignals, seizure detection, and forecasting DOI Creative Commons
Gadi Miron, Mustafa Halimeh, Jesper Jeppesen

et al.

Epilepsia, Journal Year: 2024, Volume and Issue: unknown

Published: June 5, 2024

Abstract Wearable devices have attracted significant attention in epilepsy research recent years for their potential to enhance patient care through improved seizure monitoring and forecasting. This narrative review presents a detailed overview of the current clinical state art while addressing how that assess autonomic nervous system (ANS) function reflect seizures central (CNS) changes. includes description interactions between CNS ANS, including physiological epilepsy‐related changes affecting dynamics. We first discuss technical aspects measuring biosignals considerations using ANS sensors practice. then detection forecasting studies, highlighting performance capability biomarkers. Finally, we address field's challenges provide an outlook future developments.

Language: Английский

Citations

3
Real-world epilepsy monitoring with ultra long-term subcutaneous EEG: a 15-month prospective study DOI Creative Commons
Pedro F. Viana, Jonas Duun‐Henriksen, Andrea Biondi

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

Novel subcutaneous electroencephalography (sqEEG) systems enable prolonged, near-continuous cerebral monitoring in real-world conditions. Nevertheless, the feasibility, acceptability and overall clinical utility of these remains unclear. We report on longest observational study using ultra long-term sqEEG to date.

Language: Английский

Citations

3
Real-Time Seizure Detection Using Behind-the-Ear Wearable System DOI

Jamie Lehnen,

Pooja Venkatesh,

Zhuoran Yao

et al.

Journal of Clinical Neurophysiology, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 20, 2024

Introduction: This study examines the usability and comfort of a behind-the-ear seizure detection device called brain (BrainSD) that captures ictal electroencephalogram (EEG) data using four scalp electrodes. Methods: is feasibility study. Thirty-two patients admitted to level 4 Epilepsy Monitoring Unit were enrolled. The subjects wore BrainSD standard 21-channel video-EEG simultaneously. Epileptologists analyzed EEG signals collected by validated it confirm its accuracy. A poststudy survey was completed each participant evaluate device. In addition, focus group UT Southwestern epileptologists held discuss features they would like see in home EEG-based such as BrainSD. Results: total, captured 11 14 seizures occurred while being worn. All on had focal onset, with three becoming bilateral tonic-clonic one subclinical status. worn for an average 41 hours. showed most users found comfortable, easy-to-use, stated be interested expressed similar interest Conclusions: Brain able detect Its comfort, ease-of-use, ability numerous types make acceptable at-home from both patient provider perspective.

Language: Английский

Citations

2
Training size predictably improves machine learning-based epileptic seizure forecasting from wearables DOI Creative Commons
Mustafa Halimeh, Michele Jackson, Tobias Loddenkemper

et al.

Neuroscience Informatics, Journal Year: 2024, Volume and Issue: 5(1), P. 100184 - 100184

Published: Dec. 9, 2024

Language: Английский

Citations

2