aceBeyond BMI: Deep Learning Segmentation-Driven CT Reveals Body Composition Changes after Metabolic and Bariatric Surgery
Background BMI is the primary metric used to evaluate outcomes of metabolic and bariatric surgery (MBS), but it does not distinguish tissue compartments or quantify visceral adiposity (VAT), a key determinant of cardiometabolic risk. We evaluated the relationship between BMI and VAT and characterized compartment-specific remodeling after MBS using artificial intelligence-enabled CT segmentation. Study design A retrospective analysis of prospectively collected abdominal CT scans was performed at a single tertiary…
The surgery of the subperitoneal tissue : read in the Section of Surgery, at the Annual Meeting of the British Medical Association in Carlisle, July, 1896 by Anderson, William, 1842-1900 Royal College of Surgeons of England. Public domain
Researchers retrospectively analyzed prospectively collected abdominal CTs using Comp2Comp, a validated deep learning pipeline that automatically segments visceral adipose tissue, subcutaneous adipose tissue, and skeletal muscle. They studied 435 adults with BMI >=25 for baseline BMI-VAT relationships and 39 metabolic and bariatric surgery patients with 151 scans followed up to 89 months to track compartment changes.
The work matters because BMI is the standard outcome metric after bariatric surgery but does not distinguish fat compartments, while visceral adiposity drives cardiometabolic risk. Showing that AI-enabled CT can quantify sustained VAT loss beyond BMI suggests a more precise way to monitor surgical response, though the small longitudinal sample, single-center retrospective design, and lack of outcome validation leave uncertainty about clinical utility.
- Population cohort included 435 adults with BMI greater than or equal to 25 kg/m2 undergoing CT for clinical indications to assess baseline BMI-VAT association.
- Longitudinal MBS cohort included 39 patients with complete follow-up contributing 151 CT studies with follow-up to 89 months.
- In population cohort, VAT was moderately correlated with BMI (r = 0.36).
- Study used Comp2Comp, a validated deep learning pipeline for automated segmentation of visceral adipose tissue, subcutaneous adipose tissue, and skeletal muscle.
Deep learning CT segmentation using Comp2Comp enabled compartment-specific measurement of visceral adipose tissue, subcutaneous fat, and muscle, revealing sustained visceral fat reduction after metabolic and bariatric surgery that BMI alone does not capture.
The rundown
The study analyzed two groups from a single tertiary center: 435 adults with BMI >=25 undergoing clinical CT to examine BMI-VAT correlation, and 39 bariatric surgery patients with 151 CTs tracked up to 89 months for temporal changes in BMI, VAT, and muscle. Segmentation was automated via Comp2Comp for VAT, subcutaneous adipose tissue, and skeletal muscle.
Results showed only moderate correlation between BMI and VAT at baseline (r=0.36), and the authors concluded BMI incompletely reflects postoperative tissue remodeling, particularly sustained VAT reduction. They position AI-enabled volumetric CT as proof-of-concept for compartment-specific assessment, with need for prospective validation for cardiometabolic prediction.
Sources
- Peer-reviewedJournal of the American College of Surgeons2026-07-16
How should this claim be treated?
The debate