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

Those who were on the path to retirement were more likely to commit errors and less likely to attend tumor boards.

Of 1,000 patients injected with corticosteroids under fluoroscopic guidance at an academic medical center over a 4½-year period, only 10 experienced serious complications within a year. 

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.

The interventional radiology practice of an academic medical center has identified four challenges to securing informed consent from patients or their medical decision-makers. 

"Inattention blindness bias" causes radiologists to unintentionally overlook what could be considered an obvious or significant finding.

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.

Radiologists have developed an abdominal x-ray protocol that quantifies patient “thickness” to yield images of high quality while reducing radiation dose by 80%.