Chapter 8 - Hashing-based large-scale medical image retrieval for computer-aided diagnosis

Abstract

With the ever-increasing amount of annotated medical data, large-scale, data-driven methods provide the promise of bridging the semantic gap between images and diagnoses. Our goal is to increase the scale at which interactive systems can be effective for knowledge discovery in potentially massive databases of medical images. In particular, we investigate large-scale medical image retrieval techniques, with histopathological images as the use case. In this chapter, we present hashing methods to bridge the semantic gap. With a small amount of supervised information, our method can compress a high-dimensional image feature vector into tens of bits with informative signatures preserved, and these binary codes are indexed into a hash table for real-time retrieval. We validate the hashing-based image retrieval framework on several thousands of images of breast microscopic tissues for both image classification and retrieval. Our framework achieves high search accuracy and promising computational efficiency, comparing favorably with other commonly used methods.