Prediction of progression in Parkinson’s disease based on radiomics in T1-weighted MRI and α‑synuclein in cerebrospinal fluid

Motivation: By extracting a significant amount of information from clinical imaging, radiomics makes precise and effective clinical diagnosis possible with the advent of machine learning techniques. Goal(s): This study aimed to construct a radiomics and combined model that predicted disease stage and progression of Parkinson's disease (PD). Approach: The approach is extracting radiomics features from standard T1-weighted MRI and clinical features from total CSF α-syn to establish prediction model based on machine learning. Results: The radiomics-based classifier completed task excellent and the combined model outperformed the former. Impact: This study demonstrated the feasibility of selecting specific ROIs in standard T1-weighted MRI to predict the course of PD. Our work confirmed that the five brain regions under investigation will in fact alter as PD progresses, as shown by radiomics.