Abstract

Info

Title:: Diabetic retinopathy segmentation using dense dilated attention pyramid and multi-scale features

Author(s):: WANG Zhilu1; CHI Yue1; ZHOU Yatong1; SHAN Chunyan2; XIAO Zhitao3; WANG Shaoqi1; 1. School of Electronic and Information Engineering, Hebei University of Technology, Tianjin 300401, China 2. Tianjin Medical University Chu Hsien-I Memorial Hospital/Tianjin Institute of Endocrinology/Key Laboratory of Hormones and Development of the National Health Commission/Tianjin Key Laboratory of Metabolic Diseases, Tianjin 300134, China 3. School of Life Sciences, Tiangong University, Tianjin 300387, China

Keywords:: Keywords: diabetic retinopathy dense dilated attention pyramid multi-scale feature residual module

Abstract:: Abstract: An improved U-shaped multi-lesion segmentation model, namely dense dilated attention pyramid UNet (DDAPNet), is proposed to overcome the difficulty in learning multi-scale features and address the issue of blurry boundaries in diabetic retinopathy (DR) segmentation task. DR images are treated with Patch processing to enhance the models ability to capture local lesion features. After backbone feature extraction, a redesigned dense dilated attention pyramid module is introduced to expand the receptive field and address the issue of blurry lesion boundaries and simultaneously, pyramid split attention module is used for feature enhancement and then, the features output by the two modules are fused. Additionally, an improved residual attention module is embedded within skip connections to reduce interference from shallow redundant information. The joint validation on DDR dataset and real dataset from a specific hospital shows that compared with the original model, DDAPNet model improves the Dice similarity coefficient for segmentations of microaneurysms, hemorrhages, soft exudates and hard exudates by 4.31%, 2.52%, 3.39% and 4.29%, respectively, and increases mean intersection over union by 1.80%, 2.24%, 4.28% and 1.98%, respectively. The proposed model makes the segmentation of lesion edges smoother and more continuous, notably enhancing the segmentation performance for conditions like soft exudates in retinal lesions.