Deep Representation Learning of Wrist-Worn Sensor Signals for Latent Motor Abnormality Scoring in Parkinson's Disease
- 1Iran University of Science and Technology
Abstract
Objective quantification of motor abnormality in Parkinsons disease (PD) remains challenging because conventional clinical assessments are episodic, observer-dependent, and based on coarse ordinal ratings. Wrist-worn wearable sensors provide a scalable opportunity to capture task-specific motor patterns; however, most existing approaches focus on discrete classification rather than deriving continuous latent markers of motor abnormality. In this paper, a multi-stream deep representation network is introduced to derive a latent motor abnormality score using wrist-worn inertial sensor signals. The proposed network includes a local feature extractor based on one-dimensional convolutional layers, a global feature extractor based on Transformer layers, and an embedding layer that maps each task-specific signal into a 64-dimensional embedding vector. A new training procedure is proposed based on a combination of supervised contrastive learning and center loss to cluster the embeddings of PD patients and healthy control (HC) subjects, and to build a latent motor abnormality score based on the average embedding distance from the centroids of PD and HC training embeddings. Across five-fold subject-level cross-validation, the proposed method was assessed for PD/HC classification and achieved an accuracy of 83.10% {+/-} 3.15%, balanced accuracy of 85.89% {+/-} 3.93%, precision of 97.02% {+/-} 2.33%, and ROC-AUC of 0.917. The motor abnormality score showed clear group separation, with healthy controls generally obtaining negative scores and PD subjects obtaining positive scores. Statistical analysis confirmed a significant difference between groups (F = 222.99, p < 0.001), with a substantial effect size ({superscript 2} = 0.3871). Moreover, exploratory analyses showed that the score tended to increase with higher non-motor symptom burden and longer disease duration, suggesting that the learned latent representation captured disease-related motor abnormalities.
Related
- Lymphocyte-predominant inflammation refines lung histopathology scoring in preclinical respiratory viral infection: support from deep learning analysis
- Machine Learning-Based Fatigue Detection from EEG Signals: A Methodological Pilot Framework with Potential for Nursing Applications
- Machine Learning Detection of Peritoneal Entry from Vibroacoustic Signals During Veress Needle Insertion
- Atypical Cerebellar Vermis Signal Abnormality in Wernicke Encephalopathy
- A deep learning-based silhouette gait detection framework for fall risk classification in Parkinson’s disease
BibTeX
@misc{mohtavipour2026deep,
title = {{Deep Representation Learning of Wrist-Worn Sensor Signals for Latent Motor Abnormality Scoring in Parkinson's Disease}},
author = {S. M. Mohtavipour},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.03.26362173},
url = {https://www.medrxiv.org/content/10.64898/2026.09.03.26362173v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Deep Representation Learning of Wrist-Worn Sensor Signals for Latent Motor Abnormality Scoring in Parkinson's Disease
AU - S. M. Mohtavipour
PY - 2026
DA - 2026/09/03
PB - medRxiv
DO - 10.64898/2026.09.03.26362173
UR - https://www.medrxiv.org/content/10.64898/2026.09.03.26362173v1
AB - Objective quantification of motor abnormality in Parkinsons disease (PD) remains challenging because conventional clinical assessments are episodic, observer-dependent, and based on coarse ordinal ratings. Wrist-worn wearable sensors provide a scalable opportunity to capture task-specific motor patterns; however, most existing approaches focus on discrete classification rather than deriving continuous latent markers of motor abnormality. In this paper, a multi-stream deep representation network is introduced to derive a latent motor abnormality score using wrist-worn inertial sensor signals. The proposed network includes a local feature extractor based on one-dimensional convolutional layers, a global feature extractor based on Transformer layers, and an embedding layer that maps each task-specific signal into a 64-dimensional embedding vector. A new training procedure is proposed based on a combination of supervised contrastive learning and center loss to cluster the embeddings of PD patients and healthy control (HC) subjects, and to build a latent motor abnormality score based on the average embedding distance from the centroids of PD and HC training embeddings. Across five-fold subject-level cross-validation, the proposed method was assessed for PD/HC classification and achieved an accuracy of 83.10% {+/-} 3.15%, balanced accuracy of 85.89% {+/-} 3.93%, precision of 97.02% {+/-} 2.33%, and ROC-AUC of 0.917. The motor abnormality score showed clear group separation, with healthy controls generally obtaining negative scores and PD subjects obtaining positive scores. Statistical analysis confirmed a significant difference between groups (F = 222.99, p < 0.001), with a substantial effect size ({superscript 2} = 0.3871). Moreover, exploratory analyses showed that the score tended to increase with higher non-motor symptom burden and longer disease duration, suggesting that the learned latent representation captured disease-related motor abnormalities.
ER -