Artificial intelligence (AI)-driven automated analysis of abdominal CT images outperforms traditional biomarkers for predicting risk of heart attack and stroke, new research indicates.
“Established cardiovascular risk models rely on factors like weight and BMI that are crude surrogates of body composition,” Dr. Kirti Magudia of the University of California, San Francisco, said in a news release.
“It’s well established that people with the same BMI can have markedly different proportions of muscle and fat. These differences are important for a variety of health outcomes,” she added.
A single axial CT slice of the abdomen shows the volume of subcutaneous fat area, visceral fat area and skeletal muscle area. However, manually measuring these individual areas is “time intensive and costly,” Dr. Magudia noted in a presentation at the Radiological Society of North America (RSNA) annual meeting.
Dr. Magudia was part of a multidisciplinary team of researchers who developed a fully automated method using AI to determine body-composition metrics from abdominal CT images.
They used more than 33,000 abdominal CT exams performed on 23,136 adults, more than half of whom were free of major cardiovascular and cancer diagnoses at the time of imaging. The mean age of participants was 52 years and 57 percent of patients were women.
For each individual, the researchers calculated body-scomposition areas using the CT slice at the third lumbar-spine vertebra. Participants were divided into four quartiles based on the normalized values of subcutaneous fat area, visceral fat area and skeletal muscle area.
Among the patients without major cardiovascular and cancer diagnoses at the time of imaging, 1,560 suffered a myocardial infarction and 938 suffered a stroke within five years of their abdominal CT scan.
Statistical analysis showed that normalized visceral fat area was independently associated with future risk of MI and stroke.
“Patients with the highest amount of visceral fat area, in quartile four, had a higher risk of heart attack in the five years after having their CT,” even when adjusted for known cardiovascular risk factors (P=0.04), Dr. Magudia reported. “The group of patients with the lowest amount of visceral fat area were actually protected from stroke in the years following the abdominal CT exam.”
The risks of MI and stroke were 31 percent and 46 percent higher, respectively, in the fourth versus first quartile, while there was no link to height and weight.
Summing up, Dr. Magudia said, “fully automated and normalized body-composition analysis can be applied to large-scale research projects with a low failure rate and normalized visceral fat area are associated with subsequent heart attack and stroke.”
Pages: 1 2 | Single Page
No Responses to “AI-driven Body-composition Analysis Predicts Heart Attack and Stroke Better than BMI”