Agarwala S, Kale M, Kumar D, et al. - Researchers sought to establish a screening tool for the detection of interstitial lung disease (ILD) patterns using a deep-learning method. This study applied a fully convolutional network for the semantic segmentation of several ILD patterns. Applying multi-scale feature extraction, improved segmentation of ILD patterns was achieved. Dilated convolution was applied to keep the resolution of feature maps and to enlarge the receptive field. They assessed the proposed method on a publicly available ILD database (MedGIFT) and a private clinical research database. For quantitative evaluation of the proposed method, several metrics, such as success rate, sensitivity, and false positives per section were applied. Utilizing a deep-learning framework, automatic identification of ILD patterns in a high-resolution CT image was implemented. It was demonstrated that creation of a pre-trained model with natural images and subsequent transfer learning using a particular database gives acceptable outcomes.