Speech-Based Parkinson's Disease Classification with KNN: Achieving High Accuracy in Non-Invasive Diagnostics

Progressive neurodegenerative illness Parkinson's disease (PD) is typified by non-motor symptoms that seriously impair quality of life as well as motor symptoms including tremors, stiffness, and bradykinesia. Good management and treatment of Parkinson's disease depend on an early and precise diagnosis. used a number of machine learning techniques to divide the voice samples into groups with and without Parkinson's Disease. At an accuracy of 88.8%, the K-Nearest Neighbours (KNN) method outperformed the other models examined. Strong training and validation procedures were made possible by the large dimensionality and variety of the features in the dataset, underscoring the promise of speech signal analysis for Parkinson's disease early diagnosis and monitoring. We show that KNN is effective in managing complicated, high-dimensional biomedical information and offer a viable path for non-invasive Parkinson's disease diagnosis tools.