Towards Topological Correct Segmentation of Macular OCT from Cascaded FCNs

Abstract

Optical coherence tomography (OCT) is used to produce high resolution depth images of the retina and is now the standard of care for in-vivo ophthalmological assessment. In particular, OCT is used to study the changes in layer thickness across various pathologies. The automated image analysis of these OCT images has primarily been performed with graph based methods. Despite the preeminence of graph based methods, deep learning based approaches have begun to appear within the literature. Unfortunately, they cannot currently guarantee the strict biological tissue order found in human retinas. We propose a cascaded fully convolutional network (FCN) framework to segment eight retina layers and preserve the topological relationships between the layers. The first FCN serves as a segmentation network which takes retina images as input and outputs the segmentation probability maps of the layers. We next perform a topology check on the segmentation and those patches that do not satisfy the topology criterion are passed to a second FCN for topology correction. The FCNs have been trained on Heidelberg Spectralis images and validated on both Heidelberg Spectralis and Zeiss Cirrus images.

Deep learning based topology guaranteed surface and MME segmentation of multiple sclerosis subjects from retinal OCT

“…An FCN outputs a label map instead of a set of single pixel classifications, which is much more computationally efficient. FCNs have been used for the segmentation of retinal layers [24,25], macular cysts [26,27], and both together [28]. However, these FCN methods for retinal layer segmentation have two major drawbacks.…”

“…Previous researchers have proposed deep segmentation networks that address shape and topology requirements [25,29,30]. BenTaieb et al [29] proposed to explicitly integrate topology priors into the loss function during training.…”

“…BenTaieb et al [29] proposed to explicitly integrate topology priors into the loss function during training. Ravishankar et al [30] and He et al [25] used a second auto-encoder network to learn the segmentation shape prior. These methods can improve the shape and topology of the segmentation results, but none guarantees the output to provide a correct ordering of the segmented layers due to the pixel-wise labeling nature of FCNs.…”

Active contour method for ILM segmentation in ONH volume scans in retinal OCT

“…Deep learning based methods are an important advancement, which might lead to powerful segmentation approaches of retinal tissue [35,36,39]. Their performance is highly dependent on the training dataset [58], and thus their applicability to a broad clinical spectrum of retinal changes remains to be shown.…”

“…These methods are all based on pixel-wise labelling without using topological relationships between layers or layer shape, which can lead to errors in the segmentation. He et al [39] proposed a framework to correct topological defects by cascading two FCNs. This method was also applied and tested only on macula scans.…”