Prediction of parkinson disease using hybrid machine learning model for Healthcare Application

Parkinson's disease (PD) is progressively emerging as a substantial global health concern, impacting multiple nations and experiencing a rise in prevalence. The diagnosis primarily depends on the analysis of symptoms, with a notable impairment in mobility, coordination, and muscle control. Common indicators encompass tremors, rigidity, postponed motor function, and impaired ambulation. Parkinson's disease (PD) frequently causes voice changes, although some modifications may not be immediately noticeable. These modifications can impact respiration, vocal intonation, and volume by a maximum of ten decibels. Studies indicate that alterations in speech frequently manifest at an early stage in Parkinson's disease, prompting the use of a dataset comprising human voice data for assessment. Speech recognition presents a potentially advantageous novel methodology for evaluating and controlling Parkinson's disease. The objective of this review is to conduct a comparative analysis of machine learning techniques employed in the diagnosis of Parkinson's disease. To evaluate the health status of individuals, a dataset of speech characteristics is analysed using machine learning algorithms. Recent studies are investigating more efficient model for ensemble learning and their impact on overall performance.