A combined machine-learning and graph-based framework for the segmentation of retinal surfaces in SD-OCT volumes

Abstract

Optical coherence tomography is routinely used clinically for the detection and management of ocular diseases as well as in research where the studies may involve animals. This routine use requires that the developed automated segmentation methods not only be accurate and reliable, but also be adaptable to meet new requirements. We have previously proposed the use of a graph-theoretic approach for the automated 3-D segmentation of multiple retinal surfaces in volumetric human SD-OCT scans. The method ensures the global optimality of the set of surfaces with respect to a cost function. Cost functions have thus far been typically designed by hand by domain experts. This difficult and time-consuming task significantly impacts the adaptability of these methods to new models. Here, we describe a framework for the automated machine-learning based design of the cost function utilized by this graph-theoretic method. The impact of the learned components on the final segmentation accuracy are statistically assessed in order to tailor the method to specific applications. This adaptability is demonstrated by utilizing the method to segment seven, ten and five retinal surfaces from SD-OCT scans obtained from humans, mice and canines, respectively. The overall unsigned border position errors observed when using the recommended configuration of the graph-theoretic method was 6.45 ± 1.87 μm, 3.35 ± 0.62 μm and 9.75 ± 3.18 μm for the human, mouse and canine set of images, respectively.

Fully automated detection of diabetic macular edema and dry age-related macular degeneration from optical coherence tomography images

“…Over the past two decades, the application of image processing and computer vision to OCT image interpretation has mostly focused on the development of automated retinal layer segmentation methods [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. Segmented layer thicknesses are compared to the corresponding thickness measurements from normative databases to help identify retinal diseases [19][20][21][22].…”

Incorporation of learned shape priors into a graph-theoretic approach with application to the 3D segmentation of intraretinal surfaces in SD-OCT volumes of mice

“…In particular, the learned shape priors were incorporated into a baseline approach 18 that did not incorporate any learned cost terms and an approach that incorporated learned on-surface and in-region cost terms. 15 Thus, in the two experiments conducted, the image-feature based cost term C I was designed using:…”

“…15 The statistical comparisons were conducted using paired t-tests that were corrected for multiple comparisons.…”

“…In the original formulation, 8,9 the cost function was designed as a weighted combination of an on-surface and an inregion cost term, where the on-surface cost reflected the cost associated with a voxel lying on a particular surface and the in-region cost reflected the cost associated with a voxel lying within a particular region. While these terms can be designed by hand 9-11 one extension to this method involved using machine-learning based techniques to design the on-surface and in-region cost terms, [12][13][14][15] which has been shown to improve the segmentation accuracy. Another extension that was proposed for this graph-based method was the incorporation of shape priors 16, 17 derived from a training set.…”

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

“…Finally, while this paper was under review, a new work by Antony and colleagues was published which addressed a graph-based method for the automated segmentation of 10 retinal layer boundaries in normal mice, excluding the optic nerve head (ONH) region [17].…”