A Machine Learning Approach to Analyzing DATSCAN SBR Values for the Detection of Parkinson's Disease

Abstract

Parkinson's disease (PD) is a progressive neurodegenerative condition characterized by complex symptoms, making early diagnosis challenging. However, early detection is achievable through DaTSCAN, which evaluates brain function instead of focusing solely on anatomical details. This study aims to develop a machine-learning model to distinguish between PD and healthy controls (HC) while examining significant changes in biomarkers in PD patients. Our research focuses explicitly on the Striatal Binding Ratio (SBR) values of the putamen and caudate nucleus, located in the basal ganglia region in the brain. These regions are primarily responsible for cognition, motor skills, and executive functions. The significance of this research lies in its potential to improve early diagnosis of PD using a Random Forest algorithm, which yielded an impressive accuracy of 97%. Timely diagnosis can significantly enhance a patient's quality of life by facilitating