Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach
ABSTRACT Purpose A debilitating and poorly understood symptom of Parkinson's disease (PD) is freezing of gait (FoG), which increases the risk of falling. Clinical evaluations of FoG, relying on patients’ subjective reports and manual examinations by specialists, are unreliable, and most detecti...
Saved in:
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2025-01-01
|
Series: | Brain and Behavior |
Subjects: | |
Online Access: | https://doi.org/10.1002/brb3.70206 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832582628933369856 |
---|---|
author | Sara Abbasi Khosro Rezaee |
author_facet | Sara Abbasi Khosro Rezaee |
author_sort | Sara Abbasi |
collection | DOAJ |
description | ABSTRACT Purpose A debilitating and poorly understood symptom of Parkinson's disease (PD) is freezing of gait (FoG), which increases the risk of falling. Clinical evaluations of FoG, relying on patients’ subjective reports and manual examinations by specialists, are unreliable, and most detection methods are influenced by subject‐specific factors. Method To address this, we developed a novel algorithm for detecting FoG events based on movement signals. To enhance efficiency, we propose a novel architecture integrating a bottleneck attention module into a standard bidirectional long short‐term memory network (BiLSTM). This architecture, adaptable to a convolution bottleneck attention–BiLSTM (CBA‐BiLSTM), classifies signals using data from ankle, leg, and trunk sensors. Finding Given three movement directions from three locations, we reduce computational complexity in two phases: selecting optimal channels through ensemble learning followed by feature reduction using attention mapping. In FoG event detection tests, performance improved significantly compared to control groups and existing methods, achieving 99.88% accuracy with only two channels. Conclusion The reduced computational complexity enables real‐time monitoring. Our approach demonstrates substantial improvements in classification results compared to traditional deep learning methods. |
format | Article |
id | doaj-art-0b86db34f66c484180d10180cd0636de |
institution | Kabale University |
issn | 2162-3279 |
language | English |
publishDate | 2025-01-01 |
publisher | Wiley |
record_format | Article |
series | Brain and Behavior |
spelling | doaj-art-0b86db34f66c484180d10180cd0636de2025-01-29T13:36:39ZengWileyBrain and Behavior2162-32792025-01-01151n/an/a10.1002/brb3.70206Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection ApproachSara Abbasi0Khosro Rezaee1Department of Biomedical Engineering Islamic Azad University of Mashhad Mashhad IranDepartment of Biomedical Engineering Meybod University Meybod IranABSTRACT Purpose A debilitating and poorly understood symptom of Parkinson's disease (PD) is freezing of gait (FoG), which increases the risk of falling. Clinical evaluations of FoG, relying on patients’ subjective reports and manual examinations by specialists, are unreliable, and most detection methods are influenced by subject‐specific factors. Method To address this, we developed a novel algorithm for detecting FoG events based on movement signals. To enhance efficiency, we propose a novel architecture integrating a bottleneck attention module into a standard bidirectional long short‐term memory network (BiLSTM). This architecture, adaptable to a convolution bottleneck attention–BiLSTM (CBA‐BiLSTM), classifies signals using data from ankle, leg, and trunk sensors. Finding Given three movement directions from three locations, we reduce computational complexity in two phases: selecting optimal channels through ensemble learning followed by feature reduction using attention mapping. In FoG event detection tests, performance improved significantly compared to control groups and existing methods, achieving 99.88% accuracy with only two channels. Conclusion The reduced computational complexity enables real‐time monitoring. Our approach demonstrates substantial improvements in classification results compared to traditional deep learning methods.https://doi.org/10.1002/brb3.70206bidirectional long short‐term memorybottleneck attention modulechannel selectionensemblingfreezing of gaitParkinson's disease |
spellingShingle | Sara Abbasi Khosro Rezaee Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach Brain and Behavior bidirectional long short‐term memory bottleneck attention module channel selection ensembling freezing of gait Parkinson's disease |
title | Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach |
title_full | Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach |
title_fullStr | Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach |
title_full_unstemmed | Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach |
title_short | Deep Learning–Based Prediction of Freezing of Gait in Parkinson's Disease With the Ensemble Channel Selection Approach |
title_sort | deep learning based prediction of freezing of gait in parkinson s disease with the ensemble channel selection approach |
topic | bidirectional long short‐term memory bottleneck attention module channel selection ensembling freezing of gait Parkinson's disease |
url | https://doi.org/10.1002/brb3.70206 |
work_keys_str_mv | AT saraabbasi deeplearningbasedpredictionoffreezingofgaitinparkinsonsdiseasewiththeensemblechannelselectionapproach AT khosrorezaee deeplearningbasedpredictionoffreezingofgaitinparkinsonsdiseasewiththeensemblechannelselectionapproach |