Full article: https://www.ajronline.org/doi/abs/10.2214/AJR.21.26882 

Early and mild stages of intersitial lung disease can be challenging to evaluate by radiography. A commercially available deep learning algorithm to detect reticular opacities in surgically confirmed interstitial lung disease was more accurate than radiologists. Gregory Lee, MD discusses a new AJR article that examines the use of a deep learning algorithm tool increased interobserver agreement for readers, as well as increased the sensitivity, specificity and accuracy of the radiologist.