Automatic segmentation of up to ten layer boundaries in SD-OCT images of the mouse retina with and without missing layers due to pathology

Abstract

Accurate quantification of retinal layer thicknesses in mice as seen on optical coherence tomography (OCT) is crucial for the study of numerous ocular and neurological diseases. However, manual segmentation is time-consuming and subjective. Previous attempts to automate this process were limited to high-quality scans from mice with no missing layers or visible pathology. This paper presents an automatic approach for segmenting retinal layers in spectral domain OCT images using sparsity based denoising, support vector machines, graph theory, and dynamic programming (S-GTDP). Results show that this method accurately segments all present retinal layer boundaries, which can range from seven to ten, in wild-type and rhodopsin knockout mice as compared to manual segmentation and has a more accurate performance as compared to the commercial automated Diver segmentation software.

Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search

“…Mathematic model based methods construct a fixed or adaptive model based on prior assumptions for the structure of the input images, and include A-scan [16,17], active contour [18][19][20][21], sparse high order potentials [22], and 2D/3D graph [23][24][25][26][27][28][29][30] based methods. Machine learning based methods formulate layer segmentation as a classification problem, where features are extracted from each layer or its boundaries and used to train a classifier (e.g.…”

“…Machine learning based methods formulate layer segmentation as a classification problem, where features are extracted from each layer or its boundaries and used to train a classifier (e.g. support vector machine, neural networks, or random forest classifiers) for determining layer boundaries [29,[31][32][33].…”

Learning layer-specific edges for segmenting retinal layers with large deformations

“…Numerous approaches have been proposed, ranging from the A-scan to graph-based retinal layer segmentation [13,14]. Graph-based approaches [15][16][17][18], especially dynamic programming approaches, have been widely employed due to their performance, limited computational complexity, and resilience to noise.…”

“…A method based on the fuzzy search space with a shape regularizer has demonstrated an improved performance with shadowed artifacts spanning a few A-scans [19]. Alternatively, iterative layer delineation has been proposed for exhaustive layer delineation in low-computational facilities [17]. Moreover, limiting the search space through traditional heuristics results in failed delineation due to the emergence of less frequent patterns.…”

3D automatic segmentation method for retinal optical coherence tomography volume data using boundary surface enhancement

“…Many OCT image segmentation methods have been developed to segment the retinal layer boundaries with varying levels of success. Fernandez et al proposed a method that used a structure tensor combined with complex di®usion¯ltering to segment seven retinal layer boundaries; 1 Mujat et al implemented a method to determine the thickness of the retinal nerve¯ber layer (RNFL) from OCT images by segmenting two boundaries using anisotropic noise suppression and deformable splines; 2 Ishikawa et al recognized retinal layer positions by peaks and valleys in an A-scan intensity pro¯le by using a mean¯lter for despeckling, which segmented¯ve layer boundaries; 3 Chiu et al presented a segmentation method that used graph theory and dynamic programming to segment seven retinal layers; 4 This method was later extended for segmentation of mouse retinal layers, 5 anterior eye images, 6 age-related macular degeneration (AMD) images 7 and diabetic macular edema images; 8 Using a similar method, Yang et al 9,10 utilized a more complex approach to calculate the weights map of graph-based method, using dualscale gradient information and shortest path search techniques to segment intra-retinal boundaries in OCT images. Yazdanpanah et al used an active contour approach for the segmentation of rodent retinas; 11 A two-step kernel-based optimization was proposed by Mishra et al 12 However, the methods in Refs.…”