Analyzing speech patterns for Parkinson’s diagnosis: Insights from ML models

Parkinson’s disease (PD) is the most ubiquitous neurological disease in the globe which affects the human neurological system. It is a primary concern to detect PD in its early stages to slow down its progress with proper treatment and foster a better quality of life for affected individuals. PD primarily impacts motor and cognitive function, recent studies also revealed that $\mathbf{9 0 \%}$ of Parkinson’s patients encounter speech difficulties in the preliminary phase of the disease. In the framework of this study, we have conducted a comparative analysis of various machine learning (ML) models, including Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), K-Nearest Neighbors (KNN), Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGB), and Principal Component Analysis (PCA), for the precise identification of PD in early phase using a speech dataset. Three different approaches are employed for classification. The XGB model performed remarkably well, with $98.30 \%$ accuracy and $96.67 \%$ precision. The results hold significant promise for enhancing early-stage PD diagnosis in healthcare centres as well as within the home environment.