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.

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.

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.

More than 40% of Americans are generally OK with the thought of AI reading their chest x-rays. Moreover, some 12.3% are very comfortable with the prospect.

Along with radiology-specific software and products, the list may include newly greenlit offerings for use in other settings that are increasingly important to multidisciplinary care.

Some 81% of the models—70 of 86 DL algorithms reported in 83 separate studies—diminished at least somewhat in diagnostic accuracy compared with their accuracy on internal datasets.

Competing to classify thyroid nodules on ultrasound images as either malignant or benign, three deep learning models have essentially drawn a tie with four radiologists.