Suitability of Dysphonia Measurements for Telemonitoring of Parkinson's Disease

TLDR

The study evaluates traditional and non‑standard dysphonia measures, including a new robust Pitch Period Entropy metric, for distinguishing healthy individuals from Parkinson’s disease patients. The authors recorded sustained phonations from 31 participants (23 with PD), selected ten uncorrelated dysphonia metrics, and used exhaustive search with a kernel support vector machine to identify four measures achieving 91.4 % classification accuracy. The combination of non‑standard dysphonia metrics with harmonics‑to‑noise ratios most effectively distinguishes healthy from PD subjects, achieving 91.4 % accuracy and demonstrating robustness to acoustic and individual variability, making them suitable for telemonitoring.

Abstract

We present an assessment of the practical value of existing traditional and non-standard measures for discriminating healthy people from people with Parkinson's disease (PD) by detecting dysphonia. We introduce a new measure of dysphonia, Pitch Period Entropy (PPE), which is robust to many uncontrollable confounding effects including noisy acoustic environments and normal, healthy variations in voice frequency. We collected sustained phonations from 31 people, 23 with PD. We then selected 10 highly uncorrelated measures, and an exhaustive search of all possible combinations of these measures finds four that in combination lead to overall correct classification performance of 91.4%, using a kernel support vector machine. In conclusion, we find that non-standard methods in combination with traditional harmonics-to-noise ratios are best able to separate healthy from PD subjects. The selected non-standard methods are robust to many uncontrollable variations in acoustic environment and individual subjects, and are thus well-suited to telemonitoring applications.