Joint retina segmentation and classification for early glaucoma diagnosis

Abstract

We propose a joint segmentation and classification deep model for early glaucoma diagnosis using retina imaging with optical coherence tomography (OCT). Our motivation roots in the observation that ophthalmologists make the clinical decision by analyzing the retinal nerve fiber layer (RNFL) from OCT images. To simulate this process, we propose a novel deep model that joins the retinal layer segmentation and glaucoma classification. Our model consists of three parts. First, the segmentation network simultaneously predicts both six retinal layers and five boundaries between them. Then, we introduce a post processing algorithm to fuse the two results while enforcing the topology correctness. Finally, the classification network takes the RNFL thickness vector as input and outputs the probability of being glaucoma. In the classification network, we propose a carefully designed module to implement the clinical strategy to diagnose glaucoma. We validate our method both in a collected dataset of 1004 circular OCT B-Scans from 234 subjects and in a public dataset of 110 B-Scans from 10 patients with diabetic macular edema. Experimental results demonstrate that our method achieves superior segmentation performance than other state-of-the-art methods both in our collected dataset and in public dataset with severe retina pathology. For glaucoma classification, our model achieves diagnostic accuracy of 81.4% with AUC of 0.864, which clearly outperforms baseline methods.

Retinal Boundary Segmentation in Stargardt Disease Optical Coherence Tomography Images Using Automated Deep Learning

“… 9 – 14 More recently, machine learning and deep learning methods have been used, including support vector machines, 15 , 16 random forest classifiers, 17 patch-based classification with convolutional neural networks 18 – 22 or recurrent neural networks, 20 , 22 semantic segmentation with fully convolutional (encoder–decoder) networks, 22 – 26 and other deep learning methods. 27 – 30 Importantly, some of these methods have been applied to OCT images from patients with age-related macular degeneration, 18 , 20 , 24 , 27 diabetic retinopathy, 11 , 25 macular telangiectasia type 2, 29 diabetic macular oedema, 13 , 23 , 24 pigment epithelium detachment, 28 glaucoma, 15 , 30 multiple sclerosis 17 , 26 retinitis pigmentosa, 31 and neurodegenerative diseases. 32 These diseases are characterized by variable thinning of the inner retinal layers (e.g., glaucoma and multiple sclerosis), thickening or cystic changes in the nuclear layers (e.g., macular telangiectasia type 2 and diabetic retinopathy) or focal disruption of the retinal pigment epithelium (RPE, e.g., age-related macular degeneration, macular telangiectasia, and pigment epithelium detachment).…”

Adversarial convolutional network for esophageal tissue segmentation on OCT images

“…The first U-net segments the several layers whereas the second one refines possible errors in the prediction, thus generating strict topologically correct segmentations [32]. Similar strategy is also presented in Wang's research, which introduces a post processing network to enforce the topology correctness [24]. The topological correction ability of these methods relies on specifically designed cost functions [30,31] or additional post-processing networks [24,32].…”

“…Although the segmentation results are promising, such methods usually suffer from large redundancy and result in more inference time [22]. A more elegant framework is pixel classifying by the fully convolutional network (FCN) [23,24]. This kind of method takes advantage of convolutional networks and uses an encoder-decoder architecture to assign each pixel to a label.…”

Dual-Stage U-Shape Convolutional Network for Esophageal Tissue Segmentation in OCT Images

“…Venhuizen et al implemented retinal thickness measurement and intraretinal cystoid fluid quantification based on U-shape FCN architecture [31]. Researchers have demonstrated that the second strategy possesses state-of-the-art performance on pixel-wise labeling [22], [23], [32] in biomedical image segmentation with limited training set. As a result, this strategy is also employed in the current work.…”