Numerical Modeling of the Major Temporal Arcade Using a Quantum Genetic Algorithm

Abstract

The Major Temporal Arcade (MTA) is the thickest vessel in the retina, which can be useful to analyze different pathologies related to the retina such as diabetic retinopathy. Consequently, its numerical modeling plays a vital role in systems that perform computer aided-diagnosis in Ophthalmology. In the present chapter, a novel method for the automatic modeling of the MTA is introduced. The method consists of the steps of automatic MTA segmentation and numerical modeling based on spline curves and the use of the Quantum genetic algorithm (QGA). In this step, the QGA is analyzed and implemented in order to determine the optimal control points on a set of previously segmented vessel pixels of the MTA in retinal fundus images. These control points are used to generate the best curve to fit the MTA through spline curves. In the experimental results, the proposed method was evaluated in terms of the Mean distance to the closest point and Hausdorff distance obtaining the average values of 9.91 and 53.32, respectively, using a test set of images. Finally, in terms of computational time, the proposed method achieved an average of 7.51 s per image, which makes it suitable for computer-aided diagnosis in ophthalmology.