GoKawiilAI model from Memorial Sloan Kettering identifies brown fat on PET-CT scans
Researchers at Memorial Sloan Kettering Cancer Center built an AI tool to automatically detect brown adipose tissue (BAT) in 18F-FDG-PET-CT scans, expanding an existing patient cohort with a new dataset spanning April 2018 to July 2024. The retrospective study linked BAT presence to clinical and demographic data, including BMI and outdoor temperature, to examine associations with cardiometabolic disease risk.
GoKawiil's interpretation of the reporting above, not reported fact.
Brown fat burns calories to generate heat and has been linked to lower rates of obesity and metabolic disease, but studying it in humans has been limited by the labor-intensive process of manually identifying it on scans. An AI-based detection method could let researchers analyze far larger imaging datasets, potentially accelerating understanding of how brown fat relates to metabolic health and disease risk.
- MSK researchers developed an AI tool to detect brown adipose tissue in PET-CT scans.
- The study expanded a prior patient cohort with new scans from 2018 to 2024.
- Findings linked BAT status to cardiometabolic disease codes, BMI and outdoor temperature data.
Source: nature.com — Gómez-Banoy, 2026-10-14
Published there as: “Artificial intelligence-enabled clinical phenotyping of human brown fat”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.