Radiological Society of North America, Nov. 30-Dec. 4

The annual meeting of the Radiological Society of North America was held from Nov. 30 to Dec. 4 in Chicago, drawing participants from around the world, including radiologists, radiation oncologists, physicists in medicine, radiologic technologists, and other health care professionals. The conference featured scientific papers from a number of subspecialties covering the newest trends in radiological research, as well as education and informatics exhibits.

In one study, Cyrus A. Raji, M.D., Ph.D., of the Mallinckrodt Institute of Radiology at the Washington University School of Medicine in St. Louis, and colleagues found that higher muscle mass and lower belly fat make for a younger-looking brain on magnetic resonance imaging (MRI).

The researchers evaluated 1,164 healthy individuals (52 percent women; mean age, 55.17 years) from four sites using whole-body MRI and T1-weighted sequences. They found that brain age is favorably influenced by larger muscle mass, and older-looking brains are related to a higher amount of hidden belly fat or visceral fat, as shown on whole-body MRI.

“These findings support the importance of strength training and control of visceral fat as strategies for improving brain health,” Raji said.

Press Release

In another study, Constance D. Lehman, M.D., Ph.D., of Harvard Medical School in Boston, and colleagues evaluated the effectiveness of a U.S. Food and Drug Administration-authorized, commercially available, artificial intelligence (AI), five-year breast cancer risk model, referred to as Clairity Breast, and found that it shows stronger five-year risk stratification than breast density.

The authors trained a deep learning model on hundreds of thousands of mammograms from multiple continents and calibrated it to produce a five-year probability of breast cancer. It was applied to more than 245,000 bilateral two-dimensional screening mammograms with five-year outcome information available. The AI-risk model was compared to breast density in women undergoing routine screening mammography.

The researchers identified a strong calibration of the model”s percent-risk outputs across clinically useful five-year thresholds. Women grouped as average risk (<1.7 percent), intermediate risk (1.7 to 3.0 percent), and high risk (>3.0 percent) had observed cancer rates of 1.3, 2.7, and 5.9 percent, respectively. In contrast, density alone showed only modest risk-group separation.