Diagnostic Imaging's Weekly Scan: September 20 — September 26

Welcome to Diagnostic Imaging’s Weekly Scan, which offers an opportunity to catch up on the most well-viewed radiology content of the past week.

In multivariable external testing, tumor structure habitat phenotypes from pre-treatment MRI revealed more than a doubling of recurrence risk for women with stage 2 and stage 3 breast cancer, according to a recently published study in European Radiology.

New research in the European Journal of Radiology showed that a deep learning model developed for predicting BI-RADS assessments from dynamic contrast-enhanced (DCE) breast MRI had over a 14 percent lower AUC than radiologist evaluation in external validation testing.

In a Radiology study involving over 700 participants who had chest CT and brain MRI, those with diabetes and lower vertebral bone mineral density (vBMD) had a nearly five percent faster total accumulation of white matter hyperintensity (WMH) in comparison to those without diabetes.

In a recent interview with Diagnostic Imaging, Shadpour Demehri, MD, a co-author of the aforementioned study, discussed connections between deep learning assessment of bone mineral density on chest CT, brain MRI data and the potential for enhanced risk stratification in patients with diabetes.

In his latest blog for Diagnostic Imaging, Filippo Cademartiri, MD, PhD, discussed the advanced imaging that photon-counting CT can bring to cardiac CT. However, he also pointed out that the enhanced precision with PCCT may actually lead to less reimbursement in a fee-for-service system that rewards volume of tests and procedures.