Detecting Parkinson’s Disease from an Online Speech-task: Observational Study (Preprint)

BACKGROUND Access to neurological care—especially for Parkinson's disease (PD)—is a rare privilege for millions of people worldwide, especially in developing countries. In 2013, there were just 1200 neurologists in India for a population of 1.3 billion; the average population per neurologist exceeds 3.3 million in Africa. On the other hand, 60,000 people are diagnosed with Parkinson's disease (PD) every year in the US alone, and similar patterns of rising PD cases — fueled mostly by environmental pollution and an aging population can be seen worldwide. The current projection of more than 12 million PD patients worldwide by 2040 is only part of the picture since more than 20% of PD patients remain undiagnosed. Timely diagnosis and frequent assessment are keys to ensure timely and appropriate medical intervention, improving the quality of life for a PD patient. OBJECTIVE In this paper, we envision a web-based framework that can help anyone, anywhere around the world record a short speech task, and analyze the recorded data to screen for Parkinson’s disease (PD). METHODS We collected data from 726 unique participants (262 PD, 38% female; 464 non-PD, 65% female; average age: 61) – from all over the US and beyond. A small portion of the data (roughly 7%) was collected in a lab setting to compare quality. The participants were instructed to utter a popular pangram containing all the letters in the English alphabet “the quick brown fox jumps over the lazy dog”. We extracted both standard acoustic features (Mel Frequency Cepstral Coefficients (MFCC), jitter and shimmer variants) and deep learning-based features from the speech data. Using these features, we trained several machine learning algorithms. We also applied model interpretation techniques like SHAP (SHapley Additive exPlanations) to find out the importance of each feature in determining the model’s output. RESULTS We achieved 0.75 AUC (Area Under the Curve) performance on determining presence of self-reported Parkinson’s disease by modeling the standard acoustic features through the XGBoost – a gradient-boosted decision tree model. Further analysis reveals that the widely used MFCC features and a subset of previously validated dysphonia features designed for detecting Parkinson’s from verbal phonation task (pronouncing ‘ahh’) influence the model’s decision most. CONCLUSIONS Our model performed equally well on data collected in controlled lab environment as well as ‘in the wild’ across different gender and age groups. Using this tool, we can collect data from almost anyone anywhere with a video/audio enabled device, contributing to equity and access in neurological care.