Exploiting Evolutionary Approaches for Content-Based Medical Image Retrieval

Content-based image retrieval can be applied to assist radiologists to improve the efficiency and accuracy of interpreting the images. However, it presents some intrinsic problems. The two main problems are the so-called semantic gap that occurs due to the semantic interpretation of an image is still far to be reach, because it is based on the user´s perception about the image. The other one is the dimensionality curse which leads to high dimensional feature vectors used to represent an image, where many of these features present some correlation. To mitigate these problems the paper presents a novel framework for content-based medical image retrieval joining feature selection techniques and image descriptors with optimization methods. It is capable to not only capture the user intention, but also to tune the feature selection process through the optimization method according to each user.