Fully Automated and Standardized Abdominal CT Body Composition Analysis Promises to Augment Traditional Cardiovascular Risk Prediction Models

According to ARRS American Journal of Radiology (RDA), fully automated and standardized body composition analysis of abdominal CT promises to augment traditional cardiovascular risk prediction models.

“Visceral fat area from a fully automated and standardized analysis of abdominal CT scans predicts subsequent myocardial infarction or stroke in black and white patients, independent of traditional weight measurements, and should be considered a complement to BMI in risk models,” wrote first author Kirti. Magudia, MD, PhD, currently from the Department of Radiology at Duke University School of Medicine.

Dr. Magudia and colleagues’ retrospective study included 9,752 outpatients (5,519 women, 4,233 men; 890 self-reported blacks, 8,862 self-reported whites; mean age, 53.2 years) who underwent abdominal CT scans of routine at Brigham and Women’s Hospital or Massachusetts General Hospital. from January to December 2012, without a major cardiovascular or oncological diagnosis within 3 months of the examination. A fully automated deep learning body composition analysis was performed at the L3 vertebral level to determine three body composition zones: skeletal muscle zone, visceral fat zone, and subcutaneous fat zone. Subsequent myocardial infarction or stroke was established via electronic health records.

Ultimately, after normalizing for age, sex, and race, area of ​​visceral fat derived from routine CT was associated with risk of myocardial infarction (HR 1.31 [1.03-1.67]p=0.04 for overall effect) and stroke (HR 1.46 [1.07-2.00], p = 0.04 for overall effect) in multivariate models in black and white patients; normalized weight, BMI, skeletal muscle area, and subcutaneous fat area were not.

Noting that their large study demonstrates a pipeline for the analysis of body composition and age-, sex-, and race-specific reference values ​​to add prognostic utility to clinical practice, “we anticipate that the analysis fully automated body composition using machine learning could be widely adopted to exploit the latent value of routine imaging studies,” the authors of this RDA article concluded.

Source of the story:

Material provided by American Roentgen Ray Society. Note: Content may be edited for style and length.

Comments are closed.