Enhancing the Quality of Layer Detection in Tomographic Images of the Eye

摘要

Most of the software packages offered these days, delivered together with tomographs generalizes the results obtained, due to optimizing the time of analysis. In consequence, the analysis is quick and less accurate. The authors propose an algorithm that would improve the quality and accuracy of results obtained from methods of automatic detection of individual layers of the posterior segment of the eye, by the three-dimensional and high-speed spectral domain optical coherence tomography (SD-OCT) instrument. A new method has been proposed to enhance the results obtained from automatic analyses of layers (retinal pigment epithelium, inner and outer segments of photoreceptors etc.) in the posterior segment of the eye. The proposed method is a new one, suggested by the authors, for contextual analysis of pixels. The proposed method is fully automatic, while the measurable improvement of results is possible for application in case of 2D images, and on their basis the reconstructed 3D images. The algorithm has been proposed and implemented in the Matlab environment and C language.