Automated deep studying evaluation of abdominal CT pictures produces a more exact measurement of physique composition and predicts main cardiovascular occasions, such as heart attack and stroke, higher than general weight or physique mass index (BMI), based on a study.
The study was offered on the annual assembly of the Radiological Society of North America (RSNA).
“Established cardiovascular risk models rely on factors like weight and BMI that are crude surrogates of body composition,” mentioned Kirti Magudia, MD, PhD, an abdominal imaging and ultrasound fellow on the University of California San Francisco. “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.”
Unlike BMI, which is predicated on peak and weight, a single axial CT slice of the stomach visualizes the amount of the subcutaneous fat space, visceral fat space and skeletal muscle space. However, manually measuring these particular person areas is time-intensive and dear.
As a radiology resident at Brigham and Women’s Hospital in Boston, Dr Magudia was a part of a multidisciplinary crew of researchers, together with radiologists, a knowledge scientist and biostatistician, who developed a absolutely automated methodology utilizing deep studying – a kind of synthetic intelligence (AI) – to find out physique composition metrics from abdominal CT pictures.
“Abdominal CT scans that are routinely performed provide a more granular way of looking at body composition, but we’re not currently taking advantage of it,” Dr Magudia mentioned.
The study cohort was derived from the 33,182 abdominal CT outpatient exams carried out on 23,136 sufferers at Partners Healthcare in Boston in 2012. The researchers recognized 12,128 sufferers who had been freed from main cardiovascular and most cancers diagnoses on the time of imaging. The imply age of the sufferers was 52 years, and 57% of sufferers had been girls.
The researchers chosen the L3 CT slice (from the third lumbar backbone vertebra) and calculated physique composition areas for every affected person. Patients had been then divided into 4 quartiles based mostly on the normalized values of the subcutaneous fat space, visceral fat space and skeletal muscle space.
In this retrospective study, it was decided which of those 12,128 sufferers had a myocardial infarction (heart attack) or stroke inside 5 years after their index abdominal CT scan. The researchers discovered 1,560 myocardial infarctions and 938 strokes occurred on this study group.
Statistical evaluation demonstrated that visceral fat space was independently related to a future heart attack and stroke. BMI was not related to a heart attack or stroke.
“The group of patients with the highest proportion of visceral fat area were more likely to have a heart attack, even when adjusted for known cardiovascular risk factors,” mentioned Dr. Magudia. “The group of patients with the lowest amount of visceral fat area were protected against stroke in the years following the abdominal CT exam.”
“These results demonstrate that precise measures of body muscle and fat compartments achieved through CT outperform traditional biomarkers for predicting risk for cardiovascular outcomes,” she added.
According to Dr Magudia, this work demonstrates that absolutely automated and normalized physique composition evaluation might now be utilized to giant-scale analysis initiatives.
“This work shows the promise of AI systems to add value to clinical care by extracting new information from existing imaging data,” Dr Magudia mentioned. “The deployment of AI systems would allow radiologists, cardiologists and primary care doctors to provide better care to patients at a minimal incremental cost to the health care system.”
This paper is the recipient of an RSNA 2020 Trainee Research Prize.
Co-authors are Christopher P. Bridge, D.Phil., Camden P. Bay, PhD, Florian J. Fintelmann, MD, Ana Babic, PhD, Katherine P. Andriole, PhD, Brian M. Wolpin, MD, and Michael H. Rosenthal, MD, PhD.
(This story has been printed from a wire company feed with out modifications to the textual content. Only the headline has been modified.)