Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data

Abstract Background Classifying and predicting Parkinson's disease (PD) is challenging because of its diverse subtypes based on severity levels. Currently, identifying objective biomarkers associated with disease severity that can distinguish PD subtypes in clinical trials is necessary. This st...

Full description

Saved in:
Bibliographic Details
Main Authors: Hwayoung Park, Changhong Youm, Sang-Myung Cheon, Bohyun Kim, Hyejin Choi, Juseon Hwang, Minsoo Kim
Format: Article
Language:English
Published: BMC 2025-06-01
Series:Journal of NeuroEngineering and Rehabilitation
Subjects:
Online Access:https://doi.org/10.1186/s12984-025-01648-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850224014654439424
author Hwayoung Park
Changhong Youm
Sang-Myung Cheon
Bohyun Kim
Hyejin Choi
Juseon Hwang
Minsoo Kim
author_facet Hwayoung Park
Changhong Youm
Sang-Myung Cheon
Bohyun Kim
Hyejin Choi
Juseon Hwang
Minsoo Kim
author_sort Hwayoung Park
collection DOAJ
description Abstract Background Classifying and predicting Parkinson's disease (PD) is challenging because of its diverse subtypes based on severity levels. Currently, identifying objective biomarkers associated with disease severity that can distinguish PD subtypes in clinical trials is necessary. This study aims to address the clinical applicability and heterogeneity of PD using PD severity subtypes classification and digital biomarker development by combining objective multimodal data with machine learning (ML) approaches. Methods We analyzed datasets that combine clinical characteristics, physical function and lifestyle data, gait parameters in motion analysis systems, and wearable sensors collected from persons with PD (n = 102) to perform clustering for subtype classification. Results We identified three PD severity subtypes, each exhibiting different patterns of clinical severity, with the severity increasing as it progressed from clusters 1 to 3. We found significant mutual information between all/single modalities and the unified PD rating scale scores, identifying potential modalities with high feature importance using ML. Among all modalities, the principal components of gait parameters derived from wearable sensors were identified as the most associated indicators of PD severity. A model utilizing the first principal component of the left and right ankle achieved perfect classification with an area under the curve of 1.0, accurately distinguishing clinically severe subtypes from mild subtypes of PD. These findings suggest that gait features in both ankles can reflect asymmetry factors associated with PD severity subtypes, which contributes to high classification performance. Conclusions Digital biomarkers obtained from wearable sensors attached bilaterally to body segments demonstrate potential for classifying PD severity subtypes and tracking disease progression. Our findings emphasized the clinical value of sensor-based gait analysis in PD management, which suggested its integration into personalized monitoring systems and therapeutic interventions for persons with PD.
format Article
id doaj-art-fb703c61d76c4e4794cc87cffbcd4d63
institution OA Journals
issn 1743-0003
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series Journal of NeuroEngineering and Rehabilitation
spelling doaj-art-fb703c61d76c4e4794cc87cffbcd4d632025-08-20T02:05:45ZengBMCJournal of NeuroEngineering and Rehabilitation1743-00032025-06-0122111810.1186/s12984-025-01648-2Using machine learning to identify Parkinson’s disease severity subtypes with multimodal dataHwayoung Park0Changhong Youm1Sang-Myung Cheon2Bohyun Kim3Hyejin Choi4Juseon Hwang5Minsoo Kim6Biomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityBiomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityDepartment of Neurology, School of Medicine, Dong-A UniversityBiomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityBiomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityBiomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityBiomechanics Laboratory, Department of Healthcare and Science, College of Health Sciences, Dong-A UniversityAbstract Background Classifying and predicting Parkinson's disease (PD) is challenging because of its diverse subtypes based on severity levels. Currently, identifying objective biomarkers associated with disease severity that can distinguish PD subtypes in clinical trials is necessary. This study aims to address the clinical applicability and heterogeneity of PD using PD severity subtypes classification and digital biomarker development by combining objective multimodal data with machine learning (ML) approaches. Methods We analyzed datasets that combine clinical characteristics, physical function and lifestyle data, gait parameters in motion analysis systems, and wearable sensors collected from persons with PD (n = 102) to perform clustering for subtype classification. Results We identified three PD severity subtypes, each exhibiting different patterns of clinical severity, with the severity increasing as it progressed from clusters 1 to 3. We found significant mutual information between all/single modalities and the unified PD rating scale scores, identifying potential modalities with high feature importance using ML. Among all modalities, the principal components of gait parameters derived from wearable sensors were identified as the most associated indicators of PD severity. A model utilizing the first principal component of the left and right ankle achieved perfect classification with an area under the curve of 1.0, accurately distinguishing clinically severe subtypes from mild subtypes of PD. These findings suggest that gait features in both ankles can reflect asymmetry factors associated with PD severity subtypes, which contributes to high classification performance. Conclusions Digital biomarkers obtained from wearable sensors attached bilaterally to body segments demonstrate potential for classifying PD severity subtypes and tracking disease progression. Our findings emphasized the clinical value of sensor-based gait analysis in PD management, which suggested its integration into personalized monitoring systems and therapeutic interventions for persons with PD.https://doi.org/10.1186/s12984-025-01648-2Parkinson's diseaseSeverity subtypeMultimodal dataMachine learningClusteringDigital biomarker
spellingShingle Hwayoung Park
Changhong Youm
Sang-Myung Cheon
Bohyun Kim
Hyejin Choi
Juseon Hwang
Minsoo Kim
Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
Journal of NeuroEngineering and Rehabilitation
Parkinson's disease
Severity subtype
Multimodal data
Machine learning
Clustering
Digital biomarker
title Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
title_full Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
title_fullStr Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
title_full_unstemmed Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
title_short Using machine learning to identify Parkinson’s disease severity subtypes with multimodal data
title_sort using machine learning to identify parkinson s disease severity subtypes with multimodal data
topic Parkinson's disease
Severity subtype
Multimodal data
Machine learning
Clustering
Digital biomarker
url https://doi.org/10.1186/s12984-025-01648-2
work_keys_str_mv AT hwayoungpark usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT changhongyoum usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT sangmyungcheon usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT bohyunkim usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT hyejinchoi usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT juseonhwang usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata
AT minsookim usingmachinelearningtoidentifyparkinsonsdiseaseseveritysubtypeswithmultimodaldata