Researchers have demonstrated a deep learning model that can correct course for breast radiologists who otherwise may have erroneously deemed tissue dense in screening exams.

Post-treatment changes may mask some of the subtle, early signs of recurring breast cancer on traditional mammography, an Academic Radiology editorial explains.

Computer-aided detection boosted by AI has often proven superior to traditional CAD over the past decade, yet the “new way” has been slow to win broad adoption.

Assessing more than 11,000 patients with lesions designated as BI-RADS 4, radiologists using digital breast tomosynthesis found no significant diagnostic advantages over standard 2D mammography.

Imaging features were also examined as predictive time to resolution factors, with cortical thickness of the nodes being of particular interest.

Providers screening women with dense breasts and benign breast disease should consider individualized mammogram protocols for these patients, authors of the study suggested.

The FDA has OK’d two subsidiaries of Los Angeles-based RadNet to sell medical AI software—one product for diagnosing breast cancer, the other for streamlining MRI prostate reporting workflows. 

Combining ensemble AI models with reads from breast radiologists of mixed experience levels can help health systems consistently diagnose malignant architectural distortion on mammography.