DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation

Segmentation of retinal vessels from fundus images is critical for diagnosing diseases such as diabetes and hypertension. However, the inherent challenges posed by the complex geometries of vessels and the highly imbalanced distribution of thick versus thin vessel pixels demand innovative solutions...

Full description

Saved in:
Bibliographic Details
Main Authors: Yongchao Duan, Rui Yang, Ming Zhao, Mingrui Qi, Sheng-Lung Peng
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/9/1454
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850137667568664576
author Yongchao Duan
Rui Yang
Ming Zhao
Mingrui Qi
Sheng-Lung Peng
author_facet Yongchao Duan
Rui Yang
Ming Zhao
Mingrui Qi
Sheng-Lung Peng
author_sort Yongchao Duan
collection DOAJ
description Segmentation of retinal vessels from fundus images is critical for diagnosing diseases such as diabetes and hypertension. However, the inherent challenges posed by the complex geometries of vessels and the highly imbalanced distribution of thick versus thin vessel pixels demand innovative solutions for robust feature extraction. In this paper, we introduce DAF-UNet, a novel architecture that integrates advanced modules to address these challenges. Specifically, our method leverages a pre-trained deformable convolution (DC) module within the encoder to dynamically adjust the sampling positions of the convolution kernel, thereby adapting the receptive field to capture irregular vessel morphologies more effectively than traditional convolutional approaches. At the network’s bottleneck, an enhanced atrous spatial pyramid pooling (ASPP) module is employed to extract and fuse rich, multi-scale contextual information, significantly improving the model’s capacity to delineate vessels of varying calibers. Furthermore, we propose a hybrid loss function that combines pixel-level and segment-level losses to robustly address the segmentation inconsistencies caused by the disparity in vessel thickness. Experimental evaluations on the DRIVE and CHASE_DB1 datasets demonstrated that DAF-UNet achieved a global accuracy of 0.9572/0.9632 and a Dice score of 0.8298/0.8227, respectively, outperforming state-of-the-art methods. These results underscore the efficacy of our approach in precisely capturing fine vascular details and complex boundaries, marking a significant advancement in retinal vessel segmentation.
format Article
id doaj-art-00109efdd9174edabcf2fae42b6db031
institution OA Journals
issn 2227-7390
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-00109efdd9174edabcf2fae42b6db0312025-08-20T02:30:46ZengMDPI AGMathematics2227-73902025-04-01139145410.3390/math13091454DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel SegmentationYongchao Duan0Rui Yang1Ming Zhao2Mingrui Qi3Sheng-Lung Peng4School of Atmosphere and Remote Sensing, Wuxi University, Wuxi 214105, ChinaSchool of Computer Science, Yangtze University, Jingzhou 434025, ChinaSchool of Computer Science, Yangtze University, Jingzhou 434025, ChinaSchool of Internet of Things Engineering, Wuxi University, Wuxi 214105, ChinaDepartment of Creative Technologies and Product Design, National Taipei University of Business, Taipei 10051, ChinaSegmentation of retinal vessels from fundus images is critical for diagnosing diseases such as diabetes and hypertension. However, the inherent challenges posed by the complex geometries of vessels and the highly imbalanced distribution of thick versus thin vessel pixels demand innovative solutions for robust feature extraction. In this paper, we introduce DAF-UNet, a novel architecture that integrates advanced modules to address these challenges. Specifically, our method leverages a pre-trained deformable convolution (DC) module within the encoder to dynamically adjust the sampling positions of the convolution kernel, thereby adapting the receptive field to capture irregular vessel morphologies more effectively than traditional convolutional approaches. At the network’s bottleneck, an enhanced atrous spatial pyramid pooling (ASPP) module is employed to extract and fuse rich, multi-scale contextual information, significantly improving the model’s capacity to delineate vessels of varying calibers. Furthermore, we propose a hybrid loss function that combines pixel-level and segment-level losses to robustly address the segmentation inconsistencies caused by the disparity in vessel thickness. Experimental evaluations on the DRIVE and CHASE_DB1 datasets demonstrated that DAF-UNet achieved a global accuracy of 0.9572/0.9632 and a Dice score of 0.8298/0.8227, respectively, outperforming state-of-the-art methods. These results underscore the efficacy of our approach in precisely capturing fine vascular details and complex boundaries, marking a significant advancement in retinal vessel segmentation.https://www.mdpi.com/2227-7390/13/9/1454retinal vessel segmentationdeformable convolutioncontextual feature fusionU-Net architecture
spellingShingle Yongchao Duan
Rui Yang
Ming Zhao
Mingrui Qi
Sheng-Lung Peng
DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
Mathematics
retinal vessel segmentation
deformable convolution
contextual feature fusion
U-Net architecture
title DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
title_full DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
title_fullStr DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
title_full_unstemmed DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
title_short DAF-UNet: Deformable U-Net with Atrous-Convolution Feature Pyramid for Retinal Vessel Segmentation
title_sort daf unet deformable u net with atrous convolution feature pyramid for retinal vessel segmentation
topic retinal vessel segmentation
deformable convolution
contextual feature fusion
U-Net architecture
url https://www.mdpi.com/2227-7390/13/9/1454
work_keys_str_mv AT yongchaoduan dafunetdeformableunetwithatrousconvolutionfeaturepyramidforretinalvesselsegmentation
AT ruiyang dafunetdeformableunetwithatrousconvolutionfeaturepyramidforretinalvesselsegmentation
AT mingzhao dafunetdeformableunetwithatrousconvolutionfeaturepyramidforretinalvesselsegmentation
AT mingruiqi dafunetdeformableunetwithatrousconvolutionfeaturepyramidforretinalvesselsegmentation
AT shenglungpeng dafunetdeformableunetwithatrousconvolutionfeaturepyramidforretinalvesselsegmentation