A stochastic approach to improve macula detection in retinal images

Abstract

In this paper, we present an approach to improve detectors used in medical image processing by fine-tuning their parameters for a certain dataset. The proposed algorithm uses a stochastic search algorithm to deal with large search spaces. We investigate the effectiveness of this approach by evaluating it on an actual clinical application. Namely, we present promising results with outperforming four state-of-the-art algorithms used for the detection of the center of the sharp vision (macula) in digital fundus images.