[Paper Review] Parkinson's Disease Detection through Vocal Biomarkers and Advanced Machine Learning Algorithms

Md Abu Sayed, S. Aarif Ahamed+6 more|arXiv (Cornell University)|Nov 9, 2023

Voice and Speech Disorders13 citations

TL;DR

The paper tests multiple advanced ML models on vocal biomarkers to detect Parkinson's, finding LightGBM achieves 96% accuracy and AUC with perfect sensitivity.

ABSTRACT

Parkinson's disease (PD) is a prevalent neurodegenerative disorder known for its impact on motor neurons, causing symptoms like tremors, stiffness, and gait difficulties. This study explores the potential of vocal feature alterations in PD patients as a means of early disease prediction. This research aims to predict the onset of Parkinson's disease. Utilizing a variety of advanced machine-learning algorithms, including XGBoost, LightGBM, Bagging, AdaBoost, and Support Vector Machine, among others, the study evaluates the predictive performance of these models using metrics such as accuracy, area under the curve (AUC), sensitivity, and specificity. The findings of this comprehensive analysis highlight LightGBM as the most effective model, achieving an impressive accuracy rate of 96% alongside a matching AUC of 96%. LightGBM exhibited a remarkable sensitivity of 100% and specificity of 94.43%, surpassing other machine learning algorithms in accuracy and AUC scores. Given the complexities of Parkinson's disease and its challenges in early diagnosis, this study underscores the significance of leveraging vocal biomarkers coupled with advanced machine-learning techniques for precise and timely PD detection.

Motivation & Objective

• Motivate early Parkinson's disease detection using vocal biomarkers.
• Evaluate a range of advanced machine-learning models for PD prediction.
• Identify which model provides the best predictive performance on vocal features.

Proposed method

• Extract and analyze vocal biomarkers from subjects with and without PD.
• Train and evaluate multiple ML algorithms: XGBoost, LightGBM, Bagging, AdaBoost, SVM, among others.
• Assess models using accuracy, AUC, sensitivity, and specificity.
• Highlight LightGBM as the top performer with detailed metric reporting.

Experimental results

Research questions

• RQ1 Can vocal biomarkers enable accurate early detection of Parkinson's disease?
• RQ2 Which advanced machine-learning algorithm yields the best predictive performance on vocal biomarkers for PD?
• RQ3 What are the comparative accuracy, AUC, sensitivity, and specificity across models?
• RQ4 Is LightGBM superior to other methods for this task?

Key findings

• LightGBM achieves 96% accuracy on PD detection using vocal biomarkers.
• LightGBM also attains a 96% AUC.
• LightGBM shows 100% sensitivity and 94.43% specificity, outperforming other models in accuracy and AUC.
• Other evaluated models include XGBoost, Bagging, AdaBoost, and SVM, with inferior performance to LightGBM.