Ha R, et al. - In this investigation, researchers assessed if convolutional neural networks could be used to predict which patients with pure atypical ductal hyperplasia (ADH) could be monitored safely vs undergoing surgery. For this convolutional neural network algorithm, 298 unique images from 149 patients were used. Data reported that the AUC value was 0.86 for the test set and aggregate sensitivity was 84.6%, specificity was 88.2%, and diagnostic accuracy was 86.7%. Using mammographic images, it is possible to use convolutional neural networks to discriminate pure atypical ductal hyperplasia from ductal carcinoma in situ. The prediction model could probably be further improved with a larger dataset.