← Research · University of California, San Francisco
A Comparison of FDG and Amyloid PET for the Deep Learning Prediction of Alzheimer's Disease
This research compares the efficacy of FDG PET (metabolic) and Amyloid PET (pathological) biomarkers in training Deep Learning models for Alzheimer's prediction. The study focuses on applying these models to data from underprivileged communities, evaluating accessibility versus predictive power for early diagnosis.
Notes
The first of these to be a machine-learning problem: two PET imaging modalities, FDG and amyloid, compared as inputs to a deep model predicting Alzheimer’s progression.
The through-line to the applied work is the question it turns on — not which model scores higher, but which measurement is worth trusting in the first place. That is the same question the case notes on this site spend most of their length on.
No public link for this one. Ask and I will send the manuscript.