GDCSeg-Net: general optic disc and cup segmentation network for multi-device fundus images

Abstract

Accurate segmentation of optic disc (OD) and optic cup (OC) in fundus images is crucial for the analysis of many retinal diseases, such as the screening and diagnosis of glaucoma and atrophy segmentation. Due to domain shift between different datasets caused by different acquisition devices and modes and inadequate training caused by small sample dataset, the existing deep-learning-based OD and OC segmentation networks have poor generalization ability for different fundus image datasets. In this paper, adopting the mixed training strategy based on different datasets for the first time, we propose an encoder-decoder based general OD and OC segmentation network (named as GDCSeg-Net) with the newly designed multi-scale weight-shared attention (MSA) module and densely connected depthwise separable convolution (DSC) module, to effectively overcome these two problems. Experimental results show that our proposed GDCSeg-Net is competitive with other state-of-the-art methods on five different public fundus image datasets, including REFUGE, MESSIDOR, RIM-ONE-R3, Drishti-GS and IDRiD.

A Comprehensive Review of Deep Learning Strategies in Retinal Disease Diagnosis Using Fundus Images

“…Xiang et al [131] presented their work on OD and OC segmentation, concentrating on improving the model performance over multiple datasets. This was achieved by introducing a multi-scale weight shared attention (MSA) module after the encoder phase which enhances the OD/OC feature extraction process and a depth-wise separable convolution (DCS) module after the decoder phase that accurately concentrates on the target features.…”

W-Net: A boundary-aware cascade network for robust and accurate optic disc segmentation

“…developed a generic OD and OC segmentation network for multi-device CFP to address these issues. 17 The authors mixed CFPs from different datasets and then trained and tested them on the mixed dataset. This approach performs well on the hybrid dataset but may not work well for images without similar distribution in the hybrid dataset.…”

IF-Net: information fusion network for meibomian gland area and atrophy area segmentation

“…CE-Net [4] uses convolution branches with several different receptive fields to improve the ability of the model to obtain multi-scale information. The multi-scale weighted shared attention module proposed by GDCSeg-Net [5] can focus on information at different scales, and can integrate feature information with channel and spatial attention mechanism at multiple scales simultaneously, which can obtain the target feature information effectively. CS 2 -Net [6] introduces a mixed attention mechanism to increase the focus on the shape of the target.…”