EDBNet: Efficient Dual-Decoder Boosted Network for Eye Retinal Exudates Segmentation

Abstract

Diabetic retinopathy (DR) is one of the most common causes of vision loss or blindness globally. Early detection of retinal eye lesions like hard exudates, soft exudates, microaneurysms, and hemorrhages is crucial to detect DR in a human eye. Therefore, accurate segmentation of lesions from eye fundus images is essential to develop efficient automated DR detection systems. This paper presented a novel hard and soft exudates lesions segmentation method called Efficient Dual-Decoder Boosted Network (EDBNet). EDBNet is composed of the following main components: 1) pre-trained ImageNet ResNet50 encoder with Atrous Spatial Pyramid Pooling (ASPP), 2) UNet decoder block with Gated Skip Connections mechanism to enhance capture more details of fundus images, 3) dual-decoder boosted to improve the performance segmentation of retinal lesion in the eye fundus images, and fusion outputs of the dual-decoder boosted to generate enhanced exudates segmentation. The effectiveness of the proposed framework is assessed on the IDRiD publicly dataset in terms of accuracy, Area Under Precision-Recall (AUPR), IOU, and Dice metrics. EDBNet obtains 99.8, 74.4, 78.0, and 87.6% of soft exudates, respectively. For hard exudates, EDBNet achieves 99.5, 85.3, 80.3, and 89.1%, respectively. The experimental results also demonstrate that EDBNet outperforms many state-of-the-art methods.

A Noval Approach based on Dual-Branch Encoder and Attention Skip Connections Decoder for Hard Exudate Segmentation

“…They also, introduced a novel loss function to address data imbalance issues. Cortés et al 12 employed gate-controlled skip connections (GSC) in the skip connection decoder, to focus on the most valuable features based on the previous level.…”

LezioSeg: Multi-Scale Attention Affine-Based CNN for Segmenting Diabetic Retinopathy Lesions in Images

“…The manual detection and segmentation of small objects, like lesions, in fundus images is a painstaking process that consumes ophthalmologists' time and effort [4]. Furthermore, it is difficult for ophthalmology professionals to recognize lesions effectively and analyze a large number of fundus images at once, due to the complicated structure of lesions, their varied sizes, differences in brightness, and their inter-class similarities with other tissues [5]. Moreover, training new workers on this kind of diagnosis based on these complicated images requires significant time, to build knowledge through regular practice [6].…”