Pose-based Tremor Classification for Parkinson's Disease Diagnosis from Video

TL;DR

This paper introduces SPAPNet, a novel video-based system for classifying Parkinson's tremors using a lightweight attention architecture, achieving high accuracy and interpretability to support clinical diagnosis.

Contribution

The paper presents the first non-intrusive, video-based Parkinson's tremor classification system utilizing a novel attention module and pyramidal architecture for improved performance and interpretability.

Findings

Achieved 90.9% balanced accuracy in tremor classification.

Outperformed existing methods in accuracy and F1-score.

Demonstrated system's potential as a low-cost PD warning tool.

Abstract

Parkinson's disease (PD) is a progressive neurodegenerative disorder that results in a variety of motor dysfunction symptoms, including tremors, bradykinesia, rigidity and postural instability. The diagnosis of PD mainly relies on clinical experience rather than a definite medical test, and the diagnostic accuracy is only about 73-84% since it is challenged by the subjective opinions or experiences of different medical experts. Therefore, an efficient and interpretable automatic PD diagnosis system is valuable for supporting clinicians with more robust diagnostic decision-making. To this end, we propose to classify Parkinson's tremor since it is one of the most predominant symptoms of PD with strong generalizability. Different from other computer-aided time and resource-consuming Parkinson's Tremor (PT) classification systems that rely on wearable sensors, we propose SPAPNet, which only requires consumer-grade non-intrusive video recording of camera-facing human movements as input to provide undiagnosed patients with low-cost PT classification results as a PD warning sign. For the first time, we propose to use a novel attention module with a lightweight pyramidal channel-squeezing-fusion architecture to extract relevant PT information and filter the noise efficiently. This design aids in improving both classification performance and system interpretability. Experimental results show that our system outperforms state-of-the-arts by achieving a balanced accuracy of 90.9% and an F1-score of 90.6% in classifying PT with the non-PT class.