AI-enabled analysis of images meant to catch one disease can reveal others

That’s the result of a new study led by researchers at NYU Langone Health and an example of a new trend in “opportunistic screening,” in which radiologists repurpose existing medical images to diagnose beyond what the scan was originally designed for. disease.

The study, presented Dec. 4 at the annual meeting of the Radiological Society of North America (RSNA) in Chicago, features a reanalysis of a large number of abdominal scans to analyze a section of the aorta, which is performed for a variety of reasons. . Using that data from commonplace scans, the study authors then used AI to measure the amount of aortic calcium, attach a standard score to the calcification level, and use it to predict a person’s risk of having a major cardiovascular event, including a vessel blockage ( heart attack).

“Rather than relying on specialized coronary CT scans (which are rare and not always covered by insurance) to detect potentially fatal heart disease, we are looking to use artificial intelligence to help screen abdominal CT scans, These scans are performed for a variety of reasons and are designed to opportunistically capture heart disease. Bredella is the Bernard and Irene Schwartz Professor of Radiology and Clinical and Translational Science Research at NYU Grossman School of Medicine. Director.

Specifically, the researchers reviewed 3,662 CT scans in the New York area between 2013 and 2023 in mostly older men and women who underwent both abdominal scans (which captured part of the aorta) and coronary arteries. Specialized CT scan.

The researchers found that AI’s measurement of the amount of aortic calcification during abdominal scans from other causes allowed the team to accurately predict the same person’s coronary artery calcification and that person’s risk for a major cardiovascular event. The results suggest that abdominal scans alone could be used to predict heart attacks or other cardiovascular events, the authors said.

After three years of monitoring, people with aortic calcification were 2.2 times more likely to develop a serious heart attack or blockage of a blood vessel in the brain or need surgery to restore blood flow to the heart, and this did occur in 324 people, the study said. Author on study participants. The study also showed that 29% of the study participants developed early signs of calcium buildup in the arteries that had previously been thought to be absent.

The new findings support findings from a previous study published in the journal bone About using opportunistic screening to diagnose bone loss (also called osteoporosis).

In an earlier study, also with the help of a fully automated artificial intelligence algorithm, Bredella and her team from Massachusetts General Hospital and Harvard Medical School performed a secondary analysis of CT scans from 3,708 patients. Part of lung cancer screening, most of which are older patients who are current and former smokers. By analyzing lung scans while capturing images of nearby bones, researchers found signs of severe bone loss in men and women of all races and incomes.

The team reports that osteoporosis is an underdiagnosed disease in the general population and particularly among ethnic minority groups, with 38% of blacks and 55% of Asians screened. %, 56% of Hispanics, and 72% of whites have osteoporosis. The opportunistic screening tool also detected high proportions of body fat, arteriosclerosis, and fatty liver disease, all of which are associated with bone loss.

“Our study shows that opportunistic screening can help diagnose and treat osteoporosis in susceptible populations who are at greater risk of developing osteoporosis, particularly older adults and smokers,” Bredra said. “This work lays the foundation for using opportunistic screening to address gaps in osteoporosis and heart disease prevention, as well as cancer and diabetes screening.”

However, she noted that more research is needed to determine whether imaging data and analysis can adequately identify those at higher risk of severe coronary heart disease or osteoporosis early enough for treatments that could prove effective in reducing the disease. and death.

Funding support for the aortic calcification study was provided by National Institutes of Health grants UL1TR001445, R35HL144993, R01AG065330, and R01LM013344. Financial support for osteoporosis research was provided by National Institutes of Health grant K24DK109940.

In addition to Bredella, other NYU Langone researchers involved in the study published in RSNA include co-investigators Jeffrey Berger, MD; Soterios Gyftopoulos, MD, MBA, MS; and Barry Dann, MD ; Eduardo Iturat, MD; Michael Recht, MD; and Judy Chung, PhD. Another study co-investigator is Malte Westerhoff of Visage Imaging GmbH in Berlin, Germany.

Other co-investigators on the osteoporosis study include Florian Huber, MD; Katherine Bunnell; and Efren Flores, MD, of Massachusetts General Hospital and Harvard Medical School, Boston; and Perry Pickhardt, MD, of the University of Wisconsin-Madison ; and Ronald Summers, MD, PhD, from the National Institutes of Health Clinical Center in Bethesda, Maryland.