Logistic Regression-Based Detection of Parkinson’s Disease

Millions of people worldwide suffer with Parkinson's disease (PD), a neurodegenerative condition that impairs cognition and movement. For management and treatment to be effective, early detection is essential. This study uses logistic regression, a popular classification algorithm, to develop a model for Parkinson's disease detection. To differentiate between people with Parkinson's disease and healthy controls, the model uses a dataset that includes a number of speech features, such as fundamental frequency (MDVP: Fo), jitter (MDVP: Jitter, MDVP: RAP), shimmer (MDVP: Shimmer, Shimmer: DDA), noise-to-harmonics ratio (NHR), and several other acoustic features. It also uses additional dynamic parameters, such as RPDE, DFA, and D2. Using criteria for accuracy, precision, recall, and F1-score, the logistic regression model's performance is assessed, indicating its potential for PD early detection. The findings imply that logistic regression is a useful method for PD identification, providing medical practitioners in clinical settings with an easy-to-use and useful option.