Structured speech tasks reveal temporal facial variability as an objective measure in the de-novo Parkinson's disease patients.

Diminished facial expression, a hallmark symptom of Parkinson's disease (PD), significantly impacts non-verbal communication and social interactions. This study explores the use of temporal facial variability (TVV) as a objective measure for motor impairment severity in de-novo PD patients. A cohort of 42 participants performed structured speech tasks, and their facial dynamics were captured using high-resolution 3D sensors. Advanced geometric and statistical analyses, including radial curve tensors and temporal volume variation metrics, were applied to quantify facial muscle activity. Clustering and regression analyses revealed a significant association between TVV and UPDRS Part III scores, with lower facial variability correlating with greater motor impairment. The findings suggest that temporal volume variability (TVV) may serve as a promising non-invasive marker to assist in the early detection and monitoring of Parkinson's disease. This structured speech task methodology could potentially complement existing clinical tools, contributing to improved diagnostic precision.