Parkinson's Disease Detection via Self-Supervised Dual-Channel Cross-Attention on Bilateral Wrist-Worn IMU Signals

TL;DR

This paper introduces a self-supervised dual-channel cross-attention model for classifying Parkinson's disease using wrist-worn IMU sensors, achieving high accuracy with minimal labeled data and real-time deployment.

Contribution

The study presents a novel self-supervised cross-attention encoder for PD detection from bilateral wrist IMU signals, enabling effective transfer learning with limited labels.

Findings

Achieved 93.12% accuracy for HC vs. PD classification.

Achieved 87.04% accuracy for PD vs. DD classification.

Self-supervised learning improved accuracy to 93.56% with only 20% labeled data.

Abstract

Parkinson's disease (PD) is a chronic neurodegenerative disease. It shows multiple motor symptoms such as tremor, bradykinesia, postural instability, freezing of gait (FoG). PD is currently diagnosed clinically through physical exam by health-care professionals, which can be time consuming and highly subjective. Wearable IMU sensors has become a promising gateway for passive monitoring of PD patients. We propose a self-supervised cross-attention encoder that processes bilateral wrist-worn IMU signals from a public dataset called PADS, consisting of three groups, PD (Parkinson Disease), HC (Healthy Control) and DD (Differential Diagnosis) of a total of 469 subjects. We have achieved a mean accuracy of 93.12% for HC vs. PD classification and 87.04% for PD vs. DD classification. The results emphasize the clinical challenge of distinguishing Parkinson's from other neurodegenerative diseases. Self-supervised representation learning using contrastive infoNCE loss gained an accuracy of 93.56% for HC vs. PD and 92.50% for PD vs. DD using only 20% of labelled data. This demonstrates the effectiveness of our method in transfer learning for clinical use with minimal labels. The real-time applicability was tested by deploying the optimized model with a mean inference time of 48.32 ms per window on a Raspberry Pi CPU.