Abstract
Objective: This study evaluates the efficacy of voice analysis combined with machine learning (ML) techniques in enabling early, noninvasive diagnosis of Parkinson's Disease (PD).
Methods: Voice data, phonation of the vowel 'a', from three distinct datasets (two from the UCI ML Repository and one from figshare) for a total of 432 participants (278 PD patients) were analyzed. We employed four ML models - Artificial Neural Networks (ANN), Random Forest (RF), Gradient Boosting (GB), and Support Vector Machine (SVM) - alongside two ensemble methods (soft voting classifier - EVC and stacking method - ESM). The models underwent 50 iterations of evaluation, involving various data splits and 10-fold cross-validation. Comparative analysis was done using one-way ANOVA followed by Bonferroni post-hoc corrections.
Results: The ESM, SVM, and GB models emerged as the top performers, demonstrating superior performance across metrics including accuracy, sensitivity, specificity, precision, F1 score, and ROC AUC. Despite data heterogeneity and variable selection limitations, the models showed high values for all metrics.
Conclusion: Machine learning integration with voice analysis, mainly through ESM, SVM, and GB, is promising for early PD diagnosis. Using multi-source data and a large sample size enhances our findings' validity, reliability, and generalizability.
Significance: This study underscores the potential of noninvasive voice analysis combined with advanced ML to revolutionize early PD detection and pave the way for cost-effective, scalable diagnostic tools.