An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease
Walking detection in the daily life of patients with Parkinson’s disease (PD) is of great significance for tracking the progress of the disease. This study aims to implement an accurate, objective, and passive detection algorithm optimized based on an interpretable deep learning architect...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10247059/ |
_version_ | 1797660133816270848 |
---|---|
author | Min Chen Zhanfang Sun Tao Xin Yan Chen Fei Su |
author_facet | Min Chen Zhanfang Sun Tao Xin Yan Chen Fei Su |
author_sort | Min Chen |
collection | DOAJ |
description | Walking detection in the daily life of patients with Parkinson’s disease (PD) is of great significance for tracking the progress of the disease. This study aims to implement an accurate, objective, and passive detection algorithm optimized based on an interpretable deep learning architecture for the daily walking of patients with PD and to explore the most representative spatiotemporal motor features. Five inertial measurement units attached to the wrist, ankle, and waist are used to collect motion data from 100 subjects during a 10-meter walking test. The raw data of each sensor are subjected to the continuous wavelet transform to train the classification model of the constructed 6-channel convolutional neural network (CNN). The results show that the sensor located at the waist has the best classification performance with an accuracy of 98.01%±0.85% and the area under the receiver operating characteristic curve (AUC) of 0.9981±0.0017 under ten-fold cross-validation. The gradient-weighted class activation mapping shows that the feature points with greater contribution to PD were concentrated in the lower frequency band (0.5~3Hz) compared with healthy controls. The visual maps of the 3D CNN show that only three out of the six time series have a greater contribution, which is used as a basis to further optimize the model input, greatly reducing the raw data processing costs (50%) while ensuring its performance (AUC=0.9929±0.0019). To the best of our knowledge, this is the first study to consider the visual interpretation-based optimization of an intelligent classification model in the intelligent diagnosis of PD. |
first_indexed | 2024-03-11T18:26:27Z |
format | Article |
id | doaj.art-c087e3b54daf4dc3a6f388deeb4ffc60 |
institution | Directory Open Access Journal |
issn | 1558-0210 |
language | English |
last_indexed | 2024-03-11T18:26:27Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
spelling | doaj.art-c087e3b54daf4dc3a6f388deeb4ffc602023-10-13T23:00:09ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-01313937394610.1109/TNSRE.2023.331410010247059An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s DiseaseMin Chen0https://orcid.org/0000-0003-1702-1820Zhanfang Sun1Tao Xin2Yan Chen3Fei Su4https://orcid.org/0000-0002-2585-6564Department of Neurosurgery, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, ChinaDepartment of Neurology, Provincial Hospital Affiliated to Shandong First Medical University, Jinan, ChinaDepartment of Neurosurgery, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, ChinaNeurology Department, Shanghai Jiahui International Hospital, Shanghai, ChinaDepartment of Neurosurgery, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, ChinaWalking detection in the daily life of patients with Parkinson’s disease (PD) is of great significance for tracking the progress of the disease. This study aims to implement an accurate, objective, and passive detection algorithm optimized based on an interpretable deep learning architecture for the daily walking of patients with PD and to explore the most representative spatiotemporal motor features. Five inertial measurement units attached to the wrist, ankle, and waist are used to collect motion data from 100 subjects during a 10-meter walking test. The raw data of each sensor are subjected to the continuous wavelet transform to train the classification model of the constructed 6-channel convolutional neural network (CNN). The results show that the sensor located at the waist has the best classification performance with an accuracy of 98.01%±0.85% and the area under the receiver operating characteristic curve (AUC) of 0.9981±0.0017 under ten-fold cross-validation. The gradient-weighted class activation mapping shows that the feature points with greater contribution to PD were concentrated in the lower frequency band (0.5~3Hz) compared with healthy controls. The visual maps of the 3D CNN show that only three out of the six time series have a greater contribution, which is used as a basis to further optimize the model input, greatly reducing the raw data processing costs (50%) while ensuring its performance (AUC=0.9929±0.0019). To the best of our knowledge, this is the first study to consider the visual interpretation-based optimization of an intelligent classification model in the intelligent diagnosis of PD.https://ieeexplore.ieee.org/document/10247059/Parkinson’s diseasewearable sensorsdaily detectiondeep learningvisual interpretation |
spellingShingle | Min Chen Zhanfang Sun Tao Xin Yan Chen Fei Su An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease IEEE Transactions on Neural Systems and Rehabilitation Engineering Parkinson’s disease wearable sensors daily detection deep learning visual interpretation |
title | An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease |
title_full | An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease |
title_fullStr | An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease |
title_full_unstemmed | An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease |
title_short | An Interpretable Deep Learning Optimized Wearable Daily Detection System for Parkinson’s Disease |
title_sort | interpretable deep learning optimized wearable daily detection system for parkinson x2019 s disease |
topic | Parkinson’s disease wearable sensors daily detection deep learning visual interpretation |
url | https://ieeexplore.ieee.org/document/10247059/ |
work_keys_str_mv | AT minchen aninterpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT zhanfangsun aninterpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT taoxin aninterpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT yanchen aninterpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT feisu aninterpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT minchen interpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT zhanfangsun interpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT taoxin interpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT yanchen interpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease AT feisu interpretabledeeplearningoptimizedwearabledailydetectionsystemforparkinsonx2019sdisease |