Applying Machine Learning Algorithms to Analyse Parkinson’s Disease in the Age Group of 50+

Abstract

The use of machine learning techniques in telemedicine to identify Parkinson’s disease (PD) in its early stages is explored in this paper. PD is a neurodegenerative condition that primarily affects older people. Early detection is essential for effective management and treatment, but for patients, physical visits can be difficult due to mobility and communication issues. The study used the MDVP (Multidimensional Voice Program) audio data from thirty PD patients and healthy individuals to train four classification results from Support Vector Machine (SVM), Random Forest, K-Nearest Neighbour (KNN), Logistic Regression, AdaBoost classifier, Decision tree classifiers, and stacking classifier of ensemble learning technique were compared. On balanced data, the stacking classifier ensemble learning technique has 97% detection accuracy. Furthermore, these outcomes outperform recent literature-based research. The most reliable Machine Learning (ML) method for the detection of Parkinson’s disease was discovered.