Automated CT body composition analysis predicts risk of stroke and heart attack
Body composition analysis measurements derived from routine abdominal CT scans could be used to stratify patients’ risk of stroke and heart attack, new research shows.
Researchers applied a fully automated deep learning algorithm to the abdominal CT scans of nearly 10,000 outpatients from a single system from January to December of 2012 and observed visceral fat area (VFA) measurements to be associated with increased cardiovascular risk.
Corresponding author of the study, Kirti Magudia, MD, PhD, Research Affiliate with Brigham and Women's Hospital, and colleagues explained that the use of artificial intelligence could represent a means to more widespread use of body composition analyses in risk prediction.
“CT-based body composition (BC) measurements have historically been too resource intensive to analyze for widespread use and lacked robust comparison to traditional weight metrics for predicting cardiovascular risk,” they wrote.
The fully automated analysis was performed at the L3 vertebral body level to measure three BC areas—skeletal muscle area (SMA), visceral fat area (VFA) and subcutaneous fat area (SFA)—on each of the 9,752 patients’ exams. Medical records were utilized to determine whether patients suffered a stroke or heart attack at any point during the five years following their scan.
The analysis revealed that visceral fat area measurements were consistently and significantly associated with the risk of myocardial infarction and/or stroke, while normalized skeletal muscle area, subcutaneous far area, weight and BMI, were not found to be indicative of risk.
The experts concluded that their findings support a future role as an alternative to traditional BMI risk assessments, adding:
We anticipate that fully automated BC analysis using machine learning could be widely adopted to harness latent value from routine imaging studies.”
The study abstract can be viewed in the American Journal of Roentgenology.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.