Efficient VGA-Net Modification Using ConvNeXt-Tiny and GATv2 for Retinal Vessel Segmentation
Abstract
Retinal blood vessel segmentation plays a crucial role in the early detection of ocular diseases such as diabetic retinopathy, glaucoma, and macular degeneration. Existing hybrid architectures, such as VGA-Net, suffer from high computational complexity due to the VGG-16 backbone and limited attention expressiveness due to its static GAT module, yet no prior work has examined replacing both components within a patch-based graph architecture in which backbone feature quality directly conditions graph attention effectiveness. This study aims to improve the computational efficiency and topological modeling of VGA-Net by replacing VGG-16 with ConvNeXt-Tiny and substituting GAT with GATv2. The primary contribution is a 55% parameter reduction through the ConvNeXt-Tiny backbone substitution and improved vessel topology modeling through GATv2's dynamic attention mechanism, which produces fully dynamic attention coefficients per query node. Experiments were conducted on the DRIVE and STARE datasets using a consistent preprocessing pipeline, one-factor-at-a-time hyperparameter tuning, and a unified evaluation protocol across all compared methods. The proposed model achieves the lowest parameter count (5.3M) and GFLOPs (3.2443), with a competitive inference time of 61.00 ms per image, among all compared methods, while achieving competitive performance in sensitivity and topological continuity. On the DRIVE dataset, the model achieved the highest sensitivity of 0.8718 and the highest clDice of 0.8446. On the STARE dataset, the model achieved the highest sensitivity of 0.9383 and the highest clDice of 0.9055. These results demonstrate that the proposed model achieves a favorable efficiency-performance trade-off, leading to sensitivity and topological continuity at the lowest computational cost among all compared methods, at the expense of lower specificity, accuracy, Dice, and MCC relative to certain compared methods.
Downloads
References
T. A. Tani and J. Tešić, “Advancing Retinal Vessel Segmentation With Diversified Deep Convolutional Neural Networks,” IEEE Access, vol. 12, pp. 141280–141290, 2024, doi: https://doi.org/10.1109/ACCESS.2024.3467117
K.-W. Huang, Y.-R. Yang, Z.-H. Huang, Y.-Y. Liu, and S.-H. Lee, “Retinal Vascular Image Segmentation Using Improved UNet Based on Residual Module,” Bioengineering, vol. 10, no. 6, p. 722, Jun. 2023, doi: https://doi.org/10.3390/bioengineering10060722
M. Matloob Abbasi, S. Iqbal, K. Aurangzeb, M. Alhussein, and T. M. Khan, “LMBiS-Net: A lightweight bidirectional skip connection based multipath CNN for retinal blood vessel segmentation,” Sci. Rep., vol. 14, no. 1, p. 15219, Jul. 2024, doi: https://doi.org/10.1038/s41598-024-63496-9
N. Chen, Z. Zhu, W. Yang, and Q. Wang, “Progress in clinical research and applications of retinal vessel quantification technology based on fundus imaging,” Front. Bioeng. Biotechnol., vol. 12, Feb. 2024, doi: https://doi.org/10.3389/fbioe.2024.1329263
J. Liang, Y. Jiang, and H. Yan, “Skip connection information enhancement network for retinal vessel segmentation,” Med. Biol. Eng. Comput., vol. 62, no. 10, pp. 3163–3178, Oct. 2024, doi: https://doi.org/10.1007/s11517-024-03108-w
T. A. Soomro et al., “Impact of Novel Image Preprocessing Techniques on Retinal Vessel Segmentation,” Electronics (Basel)., vol. 10, no. 18, p. 2297, Sep. 2021, doi: https://doi.org/10.3390/electronics10182297
A. A. Abdulsahib, M. A. Mahmoud, H. Aris, S. S. Gunasekaran, and M. A. Mohammed, “An Automated Image Segmentation and Useful Feature Extraction Algorithm for Retinal Blood Vessels in Fundus Images,” Electronics (Basel)., vol. 11, no. 9, p. 1295, Apr. 2022, doi: https://doi.org/10.3390/electronics11091295
Y. Jalali, M. Fateh, and M. Rezvani, “VGA‐Net: Vessel graph based attentional U‐Net for retinal vessel segmentation,” IET Image Process., vol. 18, no. 8, pp. 2191–2213, Jun. 2024, doi: https://doi.org/10.1049/ipr2.13102
A. E. Ilesanmi, T. Ilesanmi, and G. A. Gbotoso, “A systematic review of retinal fundus image segmentation and classification methods using convolutional neural networks,” Healthcare Analytics, vol. 4, p. 100261, Dec. 2023, doi: https://doi.org/10.1016/j.health.2023.100261
O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Lecture Notes in Computer Science, vol. 9351, Springer, Cham, 2015, pp. 234–241. doi: https://doi.org/10.1007/978-3-319-24574-4_28
S. Xu, Z. Chen, W. Cao, F. Zhang, and B. Tao, “Retinal Vessel Segmentation Algorithm Based on Residual Convolution Neural Network,” Front. Bioeng. Biotechnol., vol. 9, p. 786425, Dec. 2021, doi: https://doi.org/10.3389/fbioe.2021.786425
Z. Li, M. Jia, X. Yang, and M. Xu, “Blood Vessel Segmentation of Retinal Image Based on Dense-U-Net Network,” Micromachines (Basel)., vol. 12, no. 12, p. 1478, Nov. 2021, doi: https://doi.org/10.3390/mi12121478
T. M. Khan, A. Robles-Kelly, S. S. Naqvi, and M. Arsalan, “Residual Multiscale Full Convolutional Network (RM-FCN) for High Resolution Semantic Segmentation of Retinal Vasculature,” in Structural, Syntactic, and Statistical Pattern Recognition (S+SSPR 2021), Lecture Notes in Computer Science, Springer, Cham, 2021, pp. 324–333. doi: https://doi.org/10.1007/978-3-030-73973-7_31
W. Jiangtao, N. I. R. Ruhaiyem, and F. Panpan, “A Comprehensive Review of U‐Net and Its Variants: Advances and Applications in Medical Image Segmentation,” IET Image Process., vol. 19, no. 1, p. e70019, Jan. 2025, doi: https://doi.org/10.1049/ipr2.70019
A. F. M. Abdun Noor, M. I. Ahasan, M. A. Khan, and G. Yang, “GeGLUNet: Structural Retinal Vessel Segmentation via Attention-Gated GeGLU and Contrastive Supervision,” n Pattern Recognition and Computer Vision, 8th Chinese Conference, PRCV 2025, Shanghai, China, October 15–18, 2025, Proceedings, Part XIV, Lecture Notes in Computer Science, Springer, Singapore, 2026, pp. 494–507. doi: https://doi.org/10.1007/978-981-95-5631-1_35
A. G. Vrahatis, K. Lazaros, and S. Kotsiantis, “Graph Attention Networks: A Comprehensive Review of Methods and Applications,” Future Internet, vol. 16, no. 9, p. 318, Sep. 2024, doi: https://doi.org/10.3390/fi16090318
P. Cibier and J.-G. Mailly, “Graph Convolutional Networks and Graph Attention Networks for Approximating Arguments Acceptability,” in Computational Models of Argument, Frontiers in Artificial Intelligence and Applications, vol. 388, IOS Press, 2024, pp. 25–36. Aug. 2024, doi: https://doi.org/10.3233/FAIA240307
L. Alzubaidi et al., “Review of deep learning: concepts, CNN architectures, challenges, applications, future directions,” J. Big Data, vol. 8, no. 1, p. 53, Mar. 2021, doi: https://doi.org/10.1186/s40537-021-00444-8
A. A. Ramadhan and M. Baykara, “A Novel Approach to Detect COVID-19: Enhanced Deep Learning Models with Convolutional Neural Networks,” Applied Sciences, vol. 12, no. 18, p. 9325, Sep. 2022, doi: https://doi.org/10.3390/app12189325
F. A. Alotaibi et al., “GPTNeXt: Biomedical Image Classification Investigations,” Diagnostics, vol. 16, no. 4, p. 581, Feb. 2026, doi: https://doi.org/10.3390/diagnostics16040581
S. Zhu, P. Wang, and K. Shen, “ProNet Adaptive Retinal Vessel Segmentation Algorithm Based on Improved UperNet Network,” Computers, Materials & Continua, vol. 78, no. 1, pp. 283–302, 2024, doi: https://doi.org/10.32604/cmc.2023.045506
Z. Han, M. Jian, and G.-G. Wang, “ConvUNeXt: An efficient convolution neural network for medical image segmentation,” Knowl. Based. Syst., vol. 253, p. 109512, Oct. 2022, doi: https://doi.org/10.1016/j.knosys.2022.109512
S. N. Yousafzai et al., “A multi-scale simplicial transformer with graph attention for facial emotion recognition,” Ain Shams Engineering Journal, vol. 16, no. 10, p. 103584, Oct. 2025, doi: https://doi.org/10.1016/j.asej.2025.103584
D. Le et al., “Deep learning for artery–vein classification in optical coherence tomography angiography,” Exp. Biol. Med., vol. 248, no. 9, pp. 747–761, May 2023, doi: https://doi.org/10.1177/15353702231181182
X. Zhang, A. Broersen, G. Van Erp, S. Pintea, and J. Dijkstra, “Continuous and complete liver vessel segmentation with graph-attention guided diffusion,” Knowl. Based. Syst., vol. 331, p. 114686, Jan. 2026, doi: https://doi.org/10.1016/j.knosys.2025.114686
J. Staal, M. D. Abramoff, M. Niemeijer, M. A. Viergever, and B. Van Ginneken, “Ridge-Based Vessel Segmentation in Color Images of the Retina,” IEEE Trans. Med. Imaging, vol. 23, no. 4, pp. 501–509, Apr. 2004, doi: https://doi.org/10.1109/TMI.2004.825627
Y. Xu, R. Quan, W. Xu, Y. Huang, X. Chen, and F. Liu, “Advances in Medical Image Segmentation: A Comprehensive Review of Traditional, Deep Learning and Hybrid Approaches,” Bioengineering, vol. 11, no. 10, p. 1034, Oct. 2024, doi: https://doi.org/10.3390/bioengineering11101034
A. D. Hoover, V. Kouznetsova, and M. Goldbaum, “Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response,” IEEE Trans. Med. Imaging, vol. 19, no. 3, pp. 203–210, Mar. 2000, doi: https://doi.org/10.1109/42.845178
M. Liu, Y. Wang, L. Wang, S. Hu, X. Wang, and Q. Ge, “IMFF-Net: An integrated multi-scale feature fusion network for accurate retinal vessel segmentation from fundus images,” Biomed. Signal Process. Control, vol. 91, p. 105980, May 2024, doi: https://doi.org/10.1016/j.bspc.2024.105980
Z. Liu, H. Mao, C.-Y. Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A ConvNet for the 2020s,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2022, pp. 11966–11976. doi: https://doi.org/10.1109/CVPR52688.2022.01167
Z. Zhang et al., “Gradient-based Parameter Selection for Efficient Fine-Tuning,” in 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2024, pp. 28566–28577. doi: https://doi.org/10.1109/CVPR52733.2024.02699
S. Woo et al., “ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders,” in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2023, pp. 16133–16142. doi: https://doi.org/10.1109/CVPR52729.2023.01548
C. Ji, “A Survey of Neural Network Optimization Algorithms,” in 2024 IEEE 4th International Conference on Data Science and Computer Application (ICDSCA), IEEE, Nov. 2024, pp. 1–7. doi: https://doi.org/10.1109/ICDSCA63855.2024.10859435
S. Shit et al., “clDice - a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2021, pp. 16555–16564. doi: https://doi.org/10.1109/CVPR46437.2021.01629
Z. Liu, M. S. Sunar, T. S. Tan, and W. H. W. Hitam, “Deep learning for retinal vessel segmentation: a systematic review of techniques and applications,” Med. Biol. Eng. Comput., vol. 63, no. 8, pp. 2191–2208, Aug. 2025, doi: https://doi.org/10.1007/s11517-025-03324-y
Q. Qin and Y. Chen, “A review of retinal vessel segmentation for fundus image analysis,” Eng. Appl. Artif. Intell., vol. 128, p. 107454, Feb. 2024, doi: https://doi.org/10.1016/j.engappai.2023.107454
D. Chicco and G. Jurman, “A statistical comparison between Matthews correlation coefficient (MCC), prevalence threshold, and Fowlkes–Mallows index,” J. Biomed. Inform., vol. 144, p. 104426, Aug. 2023, doi: https://doi.org/10.1016/j.jbi.2023.104426
D. Chicco, N. Tötsch, and G. Jurman, “The Matthews correlation coefficient (MCC) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation,” BioData Min., vol. 14, no. 1, p. 13, Feb. 2021, doi: https://doi.org/10.1186/s13040-021-00244-z
Y. Ye, C. Pan, Y. Wu, S. Wang, and Y. Xia, “MFI-Net: Multiscale Feature Interaction Network for Retinal Vessel Segmentation,” IEEE J. Biomed. Health Inform., vol. 26, no. 9, pp. 4551–4562, Sep. 2022, doi: https://doi.org/10.1109/JBHI.2022.3182471
H. Zhang et al., “BCU-Net: Bridging ConvNeXt and U-Net for medical image segmentation,” Comput. Biol. Med., vol. 159, p. 106960, Jun. 2023, doi: https://doi.org/10.1016/j.compbiomed.2023.106960
Z. Li, X. Zhang, M. Zhao, F. Shi, and W. Zhou, “Direction-guided network for retinal vessel segmentation in OCTA images,” Biomed. Signal Process. Control, vol. 103, p. 107455, May 2025, doi: https://doi.org/10.1016/j.bspc.2024.107455
Copyright (c) 2026 Billie Zandra Widiyanto, Wiharto Wiharto, Shaifudin Zuhdi
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlikel 4.0 International (CC BY-SA 4.0) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
.png)
.png)
.png)
.png)
.png)
