Content-based medical image retrieval has been an active field of research for the past few years. In content-based retrieval systems the images are first represented in terms of high dimensional features, and similarity between images are measured from distances between these features. Given the inherent complexity of analyzing medical images owing to the subtle differences and interactions between diagnoses, large-scale image retrieval is a sparsely studied field.

  We proposed a deep-learning based image retrieval system targetted for single-modality image retrieval from large-scale databases. The proposed system has the potential to be used as a part of clinical decision support system. You can read about the framework here.