Balancing practicality and complexity in neuroimaging models of Parkinson's disease progression

Reliable progression models are essential for clinical decision-making and trial design in Parkinson’s disease. We discuss linear, exponential, and sigmoidal patterns in PET and SPECT data, emphasizing the mismatch between biomarker and clinical trajectories. We propose more adaptable modeling strategies to improve patient stratification, support trial outcomes, and align imaging biomarkers with real-world disease complexity.