Joint segmentation of optic cup and optic disc using deep convolutional generative adversarial network

Abstract

Glaucoma, as one of the three major blinding ophthalmic diseases in the world, is usually accompanied by changes in the structure of the patient’s optic disc, such as optic disc atrophy and depression. Clinical ophthalmologists tend to use the cup-disc ratio as an evaluation index to realize the screening and diagnosis of glaucoma. Therefore, the accurate measurement of optic cup (OC), optic disc (OD) and other parameters is of great clinical significance for early screening of glaucoma. Inspired by game theory, this paper combines deep convolutional neural networks (DCNN) with generative adversarial networks (GAN), and proposes a model for the joint segmentation of OC and OD. Specifically, the generator is composed of a deep convolutional encoder-decoder network to jointly segment the OC and OD, and the discriminator is composed of an eight layer full convolutional neural network. The discrimination results adjust the parameters in the structure of the generator and discriminator network through back propagation to achieve the effect of autonomous learning and optimization of the model. When the proposed network and the existing networks are evaluated on the public dataset Drishti-GS1, the research results demonstrate that the proposed network can achieve a significant improvement in the overall performance.

RSAP-Net: joint optic disc and cup segmentation with a residual spatial attention path module and MSRCR-PT pre-processing algorithm

“…To demonstrate the segmentation performance of RSAP-Net, we compared the experimental results with those of FCN [20], U-Net [21], M-Net [18], POSAL [34], Join-tRCNN [38], CE-Net [32], Stack-U-Net [44], BGA-Net [45] and DCGAN' [46]. All experiments were evaluated on the Drishti-GS1 [41] dataset.The comparison results are shown in Table 4.…”

“…Although there is a good improvement in segmentation performance for both OD and OC, the number of parameters grows linearly as the model becomes more complex, significantly increasing the computational cost. DCGAN [46] combined DCNN and GAN to propose a DCGAN-based model to jointly segment OC and OD. DCGAN achieved good segmentation performance on the segmentation of OD, but performed poorly on the segmentation accuracy of OC, with an F1 score of only 0.8631.…”