Machine Learning-based Early Diagnosis and EMG-Assisted Monitoring of Parkinson’s Disorder

Parkinson’s disorder is one of the most commonly found neurodegenerative diseases that primarily affects the central nervous system. Non-motor symptoms can be observed, such as depression, anxiety, sleep disorders, and cognitive impairments, which also occur, further exacerbating this disease. Since there is no definitive test for Parkinson’s disease, it is difficult to diagnose. Doctors rely on a mental health evaluation, medical history, and physical exam, but these processes can be slow or inaccurate, especially in the early stages. Machine learning is gaining popularity in the healthcare field due to its ability to analyze complex data, identify patterns, and make predictions to improve medical decisions. This study examined methods such as K-Nearest Neighbor (KNN), Random Forest (RF), Support Vector Machine (SVM), and XGBoost (Extreme Gradient Boosting) to achieve better and improved outcomes for patients by utilizing treatment plans, ultimately enhancing the lives of individuals with Parkinson's disorder. The results indicate that early detection using EMG-based ML analysis achieves an accuracy of up to 99%, supporting timely medical intervention