Microaneurysm detection in retinal images using a rotating cross-section based model

Retinal image analysis is currently a very vivid field in biomedical image analysis. One of the most challenging tasks is the reliable automatic detection of microaneurysms (MAs). Computer systems that aid the automatic detection of diabetic retinopathy (DR) greatly rely on MA detection. In this paper, we present a method to construct an MA score map, from which the final MAs can be extracted by simple thresholding for a binary output, or by considering all the regional maxima to obtain probability scores. In contrary to most of the currently available MA detectors, the proposed one does not use any supervised training and classification. However, it is still competitive in the field, with a prominent performance in the detection of MAs close to the vasculature, regarding the state-of-the-art methods. The algorithm has been evaluated in a publicly available online challenge.