Wearable Gait Biomarkers and Explainable AI Identify High Prodromal Burden in Parkinson’s Disease

Research overview

The text below is the source article’s original abstract under a verified reuse license. cc by

Abstract Identifying Parkinson’s disease (PD) during its prodromal, non-motor phase remains a barrier to early intervention. Gait impairment represents one of the most disabling and clinically informative manifestations of PD. Subtle and sub-clinical gait impairments may occur during the prodromal phase, being potentially useful for early PD identification. This study developed an interpretable machine-learning framework using wearable inertial gait data to model and quantify high prodromal burden in PD, defined as the anamnestic presence of at least three prodromal symptoms. A total of 275 individuals with PD performed 30-m walking trials using a single lumbar inertial sensor. Thirty-five biomechanical and clinical variables were extracted, and feature selection identified five key predictors: multiscale entropy (MSE) along three axes, vertical improved harmonic ratio (iHRv), and body weight. A Random Forest classifier balanced through Conditional Tabular Generative Adversarial Network (CTGAN) augmentation achieved robust performance (ROCAUC = 0.84, PRAUC = 0.86, F1score = 0.76). Explainability analyses highlighted that increased MSE and reduced iHRv were the strongest contributors to high prodromal burden, indicating elevated gait complexity and altered spatio-temporal symmetry. These findings delineate a gait phenotype associated with prodromal burden in PD, providing mechanistic insight into early disease-related motor alterations.

Related topics