Learning of perceptual similarity from expert readers for mammogram retrieval

Image retrieval relies critically on the similarity measure used to compare a query image to a target image in a database. In this work, we explore a similarity measure for mammogram retrieval based on supervised learning from expert readers. This approach is evaluated using data collected from an observer study with a set of clinical mammograms. Our results demonstrate that the proposed machine learning approach can be used to model the notion of similarity as judged by expert readers in their interpretation of mammogram images and that it can outperform alternative similarity measures derived from unsupervised learning