Machine Learning Algorithms for Diagnosis of Parkinson's disease Based on Voice Characteristics

Bradykinesia, tremors, rigidity, and postural instability are symptoms of a degenerative neurological condition called Parkinson's disease (PD) affecting the motor movements. Deep brain stimulation, medication, and other therapies can all be used to treat theParkinson's disease symptoms, yet as of today, there is no effective treatment. Parkinson's disease must be recognized as early and precisely as feasible for effective disease treatment and the development of new therapies. The goal of this work is to create a model that can detect Parkinson's disease based on relevant clinical and demographic variables. It employs a number of machinelearning techniques, including Logistic Regression, Support Vector Machine (SVM), Gradient Boosting Classifier, K-Nearest Neighbors (KNN), Random Forest Classifier, and ensemble method, which is a voting classifier. A significant dataset that contains data on changes in speaking patterns was used to train the model. Using measures for accuracy, precision, recall, and F1-score, the machine learning model's performance is assessed and contrasted with that of the most widely used techniques for Parkinson's disease diagnosis.