• AI
• APS
• Atypical parkinsonian disorders
• Deep Radiomics
• DLR
• FDG PET
• Parkinson's disease
• PD

Parkinson’s Disease (PD) and Atypical Parkinsonian Syndromes (APS) share many phenotypic manifestations, especially in the early stages of the disease, creating clinical challenges. The ability to precisely classify observations is extremely valuable and the accurate differential diagnosis of parkinsonism has foremost therapeutic and prognostic importance.
A relevant role in the differential diagnosis of PD and APS is played by [18F]FDG PET, as metabolic patterns of regional glucose metabolism of these nosological entities are different and disease-specific. Automated artificial intelligence (AI)-based deep learning Radiomics (DLR) framework might represent an optimal imaging-biomarker that might be able to classify PD, APS and HC individuals based on [18F]FDG PET data, particularly in the early stages of the disease, when no clear metabolic patterns are detectable by qualitative and/or semi-quantitative [18F]FDG PET assessment.