Smart Detection of Parkinson’s Disease Using Random Forest and Streamlit

Parkinson’s Disease (PD) is a chronic and progressive neurodegenerative disorder that primarily affects movement and coordination, often leading to significant impairments in daily life. Early diagnosis of Parkinson’s dis- ease is crucial for effective medical interven- tion and improved patient outcomes. In re- cent years, machine learning techniques have emerged as powerful tools in the early detec- tion of complex diseases by analyzing large sets of biomedical data. This study presents the development of a web-based application that utilizes a Random Forest Classifier for the accurate detection of Parkinson’s disease. The model is trained on a publicly available dataset consisting of biomedical voice measurements, which are proven indicators of PD. To ensure model robustness and eliminate data bias, the dataset is preprocessed using Min-Max nor- malization, and key features excluding non- informative identifiers are selected. The Ran- dom Forest algorithm is chosen for its supe- rior performance in handling nonlinear data and preventing overfitting through ensemble learning. The trained model is integrated into an interactive web interface developed using Streamlit, allowing users to input biomedical voice features and receive instant predictions. The system offers an accuracy level suitable for preliminary screening, bridging the gap between clinical diagnosis and remote accessi- bility. This work demonstrates the feasibility of combining machine learning and web tech- nologies to assist healthcare professionals and patients in making informed decisions based on objective data. Future improvements may include model fine-tuning, integration with mobile platforms, and extension to other neu- rodegenerative disorders.