Region segmentation in 3-D optical coherence tomography images

Abstract

This paper describes a novel region segmentation method created to enhance spatial relationships in 3-D optical coherence tomography (OCT) images. To reduce the noise and distortion problems in low-resolution OCT images, previous work used the mean value and an enhanced-fuzzy-c-mean algorithm to cluster pixels in 2-D OCT images and find the edge between different clustered regions. To utilize more spatial relationships and to reduce computation time, the proposed method uses the mean value and a 3-D filter-based-fuzzy-c-mean algorithm to cluster pixels in 3-D OCT images and find the edge between different clustered regions. The OCT images of an artificial object used to simulate vessels are tested in the experiment, and the segmented regions of interest are reconstructed via AVIZO for 3-D display purposes.