PET Tracer Classification
June 1, 2026
·
1 min read
A deep learning model that reads a brain PET scan and identifies which radiotracer produced it.
Large PET datasets often carry missing or incorrect tracer labels, which quietly breaks downstream analysis. We trained a 2.5D ConvNeXt model to identify the tracer directly from the image and validated it across ADNI, SCAN, and an external cohort. The model supports automated quality control and curation of large imaging archives. This work is currenlty ynder review.
Authors
Postdoctoral Fellow / Research Associate
I am a Research Associate in the Department of Radiology at Mayo Clinic, developing and validating quantitative imaging biomarkers for Alzheimer’s disease and related neurodegenerative conditions.
My work includes tau PET quantification, multisite MRI harmonization, disease progression modeling, and machine learning imaging signatures of rare disorders such as multiple system atrophy.