EXPRIMIND: A MACHINE LEARNING-BASED GENE EXPRESSION ANALYSIS PLATFORM FOR PARKINSON’S DISEASE RISK PREDICTION

Abstract

Parkinson’s disease (PD) is a progressive disorder that affects the nervous system and leads to the gradual loss of dopamine-producing neurons. Detecting it early remains a challenge for doctors. This study presents ExpriMind, a framework that uses machine learning to predict the risk of Parkinson’s disease. It analyzes publicly available gene expression data from the GEO database. The framework combines gene expression analysis with predictive modeling to identify molecular patterns that differentiate PD samples from healthy ones. ExpriMind is a web application built with Streamlit, allowing users to predict risks in real time and explore gene-level insights. The results show that using machine learning on transcriptomic data can effectively identify gene expression patterns related to the disease. This approach is reproducible and scalable for assessing the risk of PD. Additionally, the framework can be used for other complex diseases, supporting precision medicine with clear and user-friendly predictive tools.