Authors:
Nadeem Akhtar,Patil Bhushan,Mahire Amit,Rahmani Naveed Akhtar,Mohammad Shakir,DOI NO:
https://doi.org/10.26782/jmcms.2026.09.00006Keywords:
Retinal Vessel Segmentation,U-Net,ASPP,CLAHE,Fundus Image.,Abstract
The diagnosis of ophthalmic diseases such as glaucoma, diabetic retinopathy, hypertension, and cardiovascular disorders can be recognized through retinal vessel segmentation from fundus images. The low contrast, noise, blurred boundaries, and substantial variations in vessel thickness make the vessel segmentation a challenging task for semantic segmentation of retinal vessels. To mitigate these limitations, an enhanced U-Net variant is proposed. The framework integrates an initial Inception block, a sequential Global-Local (GL) feature engineering module utilizing zooming and shrinking operations, parallel 1×1 and 3×3 skip paths coupled through element-wise addition, and a terminal Atrous Spatial Pyramid Pooling (ASPP) stage. Internal representations across these stages were systematically profiled using Mutual Information (MI) metrics. Quantitative evaluations demonstrate the network's strong performance across standard benchmarks, recording an mIoU of 80.87%, a Mean BF-score of 91.89%, a center-line Dice (clDice) of 80.83%, and an 84.6% vessel classification accuracy alongside a reduced 15.4% false-negative rate. Ablation trials validate the individual architectural contributions: introducing the Inception layer increases IoU from 50.84% to 59.59%, whereas employing element-wise addition over concatenation in the GL module raises clDice to 81.04% (versus 80.72%). Additionally, MI profiling confirms that average pooling and multi-rate ASPP convolutions retain vital spatial entropy and micro-vascular geometry far more effectively than standard operations.Refference:
I. Akhtar, N. and Ragavendran, U. (2019) ‘Interpretation of intelligence in CNN-pooling processes: a methodological survey’, Neural Computing and Applications [Preprint]. 10.1007/s00521-019-04296-5.
II. Bai, H. et al. (2026) ‘nnLoGoNet: a hybrid local-global network for retinal vessel segmentation with Skeleton Recall Loss’, Scientific Reports, 16(1), p. 19297. 10.1038/s41598-026-50020-4.
III. Bansal, A. et al. (2026) ‘Retinal Vessel Segmentation: A Comprehensive Review From Classical Methods to Deep Learning Advances (1982–2025)’, Advanced Intelligent Systems, n/a(n/a), p. e202501279. Available at: 10.1002/aisy.202501279.
IV. Boudegga, H. et al. (2021) ‘Fast and efficient retinal blood vessel segmentation method based on deep learning network’, Computerized Medical Imaging and Graphics, 90, p. 101902. Available at: 10.1016/j.compmedimag.2021.101902.
V. Chu Jinsong; Zheng, Wenqiang; Xu, Zhengyuan, B.Z. (2025) ‘(DA-U) 2Net: double attention U2Net for retinal vessel segmentation.’ BMC ophthalmology, 25(1), pp. 86-NA. 10.1186/s12886-025-03908-0.
VI. Das, S. et al. (2024) ‘Assessment of retinal blood vessel segmentation using U-Net model: A deep learning approach’, Franklin Open, 8, p. 100143. 10.1016/j.fraope.2024.100143.
VII. Dong Ting; Zhang, Tianyi; Wei, Lifang, H.Z. (2022) ‘Supervised learning-based retinal vascular segmentation by M-UNet full convolutional neural network’, Signal, Image and Video Processing, 16(7), pp. 1755–1761. 10.1007/s11760-022-02132-3.
VIII. Fareed, K. et al. (2025) ‘CBAM Attention Gate-Based Lightweight Deep Neural Network Model for Improved Retinal Vessel Segmentation’, International Journal of Imaging Systems and Technology, 35(1), p. e70031. 10.1002/ima.70031.
IX. Van Gelder Michael F; Dyer, Michael A; Greenwell, Thomas N; Levin, Leonard A; Wong, Rachel O; Svendsen, Clive N, R.N.C. (2022) ‘Regenerative and restorative medicine for eye disease.’, Nature medicine, 28(6), pp. 1149–1156. 10.1038/s41591-022-01862-8.
X. Godishala, A. et al. (2025) ‘Survey on retinal vessel segmentation’, Multimedia Tools and Applications, 84(11), pp. 8361–8411. 10.1007/s11042-024-19075-1.
XI. Guo Shaojia; Ge, Rui; Li, Wenjuan; Lu, Aihong; Feng, Rongzhen; Fang, Wu; Jin, Qiangguo, F.Y. (2025) ‘RetinaDA: a diverse dataset for domain adaptation in retinal vessel segmentation’, Frontiers of Computer Science, 19(8), p. NA-NA. 10.1007/s11704-024-41114-1.
XII. Hassan, G. et al. (2015) ‘Retinal Blood Vessel Segmentation Approach Based on Mathematical Morphology’, Procedia Computer Science, 65, pp. 612–622. 10.1016/j.procs.2015.09.005.
XIII. He, K. et al. (2015) ‘Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition’, IEEE Transactions on Pattern Analysis and Machine Intelligence, 37(9), pp. 1904–1916. 10.1109/TPAMI.2015.2389824.
XIV. He, K. et al. (2016) ‘Deep Residual Learning for Image Recognition’, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778. 10.1109/CVPR.2016.90.
XV. Hoover Valentina L.; Goldbaum, Michael H., A.D. K. (2000) ‘Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response’, IEEE transactions on medical imaging, 19(3), pp. 203–210. 10.1109/42.845178.
XVI. Hossen, M.F. et al. (2026) ‘A comprehensive review of U-Net architectures for medical image segmentation: emerging trends and federated learning perspectives’, Neural Computing and Applications, 38(12), p. 493. 10.1007/s00521-026-12198-6.
XVII. Islam, M.T. et al. (2023) ‘LUVS-Net: A Lightweight U-Net Vessel Segmentor for Retinal Vasculature Detection in Fundus Images’, Electronics, 12(8). 10.3390/electronics12081786.
XVIII. Issitt, A. et al. (2026) ‘Uncovering Neural Learning Dynamics through Latent Mutual Information’, Entropy, 28(1). 10.3390/e28010118.
XIX. Jebaseeli, T.J., Durai, C.A.D. and Peter, J.D. (2019) ‘Segmentation of retinal blood vessels from ophthalmologic Diabetic Retinopathy images’, Computers & Electrical Engineering, 73, pp. 245–258. 10.1016/j.compeleceng.2018.11.024.
XX. Jin Qi; Meng, Zhaopeng; Wang, Bing; Su, Ran, Q.C. (2019) ‘Construction of Retinal Vessel Segmentation Models Based on Convolutional Neural Network’, Neural Processing Letters, 52(2), pp. 1005–1022. 10.1007/s11063-019-10011-1.
XXI. Juntang Zhuang (2018) ‘LadderNet: Multi-path networks based on U-Net for medical image segmentation’ CoRR, abs/1810.0.
XXII. Kawamura, T. (2024) ‘Annual Report’:, Quarterly Journal of Marketing, 44(1), pp. 102–104. 10.7222/marketing.2024.037.
XXIII. Kim, G., Choi, S. and Yun, S.-Y. (2025) ‘A Mutual Information-Based Framework for Enhancing Graph Neural Networks on Heterophily’, IEEE Access, 13, pp. 199531–199545. 10.1109/ACCESS.2025.3624415.
XXIV. Krizhevsky, A., Sutskever, I. and Hinton, G.E. (2017) ‘ImageNet Classification with Deep Convolutional Neural Networks’, Commun. ACM, 60(6), pp. 84–90. 10.1145/3065386.
XXV. Li Fei; Yan, Fen; Meng, Jing; Guo, Yanfei; Liu, Hongjuan; Cheng, Ronghua, S.M. (2025) ‘SMFDNet: spatial and multi-frequency domain network for OCT angiography retinal vessel segmentation’, The Journal of Supercomputing, 81(4), p. NA-NA. 10.1007/s11227-025-06985-6.
XXVI. Li Manisha; Nakashima, Yuta; Nagahara, Hajime; Kawasaki, Ryo, L.V. (2020) ‘IterNet: Retinal Image Segmentation Utilizing Structural Redundancy in Vessel Networks’, 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, pp. 3645–3654. 10.1109/wacv45572.2020.9093621.
XXVII. Lin, T.-Y. et al. (2017) ‘Feature Pyramid Networks for Object Detection’, in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 936–944. 10.1109/CVPR.2017.106.
XXVIII. Mardani, K. and Maghooli, K. (2021) ‘Enhancing retinal blood vessel segmentation in medical images using combined segmentation modes extracted by DBSCAN and morphological reconstruction’, Biomedical Signal Processing and Control, 69, p. 102837. 10.1016/j.bspc.2021.102837.
XXIX. Odstrcilik Radim; Budai, Attila; Hornegger, Joachim; Jan, Jiri; Gazarek, Jiri; Kubena, Tomas; Cernosek, Pavel; Svoboda, Ondrej; Angelopoulou, Elli, J.K. (2013) ‘Retinal vessel segmentation by improved matched filtering: evaluation on a new high-resolution fundus image database’, IET Image Processing, 7(4), pp. 373–383. 10.1049/iet-ipr.2012.0455.
XXX. Oktay, O. et al. (2018) ‘Attention U-Net: Learning Where to Look for the Pancreas’, CoRR, abs/1804.0, pp. 2–5.
XXXI. Ouyang, J. et al. (2023) ‘LEA U-Net: a U-Net-based deep learning framework with local feature enhancement and attention for retinal vessel segmentation’, Complex & Intelligent Systems, 9(6), pp. 6753–6766. 10.1007/s40747-023-01095-3.
XXXII. Pizer, S.M. et al. (1987) ‘Adaptive histogram equalization and its variations’, Computer Vision, Graphics, and Image Processing, 39(3), pp. 355–368. 10.1016/S0734-189X(87)80186-X.
XXXIII. Prajna, Y. and Nath, M.K. (2022) ‘Efficient blood vessel segmentation from color fundus image using deep neural network’, Journal of Intelligent & Fuzzy Systems, 42(4), pp. 3477–3489. 10.3233/JIFS-211479.
XXXIV. Ramos-Soto, O. et al. (2021) ‘An efficient retinal blood vessel segmentation in eye fundus images by using optimized top-hat and homomorphic filtering’, Computer Methods and Programs in Biomedicine, 201, p. 105949. 10.1016/j.cmpb.2021.105949.
XXXV. Sanjeewani Arun Kumar; Akbar, Mohd; Kumar, Mohit; Yadav, Divakar, N.Y. (2024) ‘Retinal blood vessel segmentation using a deep learning method based on modified U-NET model’, Multimedia Tools and Applications, 83(35), pp. 82659–82678. 10.1007/s11042-024-18696-w.
XXXVI. Shang, Z. et al. (2024) ‘DCNet: A lightweight retinal vessel segmentation network’, Digital Signal Processing, 153, p. 104651. 10.1016/j.dsp.2024.104651.
XXXVII. Shit, S. et al. (2021) ‘clDice – a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation’, in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 16555–16564. 10.1109/CVPR46437.2021.01629.
XXXVIII. Simonyan, K. and Zisserman, A. (2015) ‘Very Deep Convolutional Networks for Large-Scale Image Recognition’, in Y. Bengio and Y. LeCun (eds) 3rd International Conference on Learning Representations, {ICLR} 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. http://arxiv.org/abs/1409.1556.
XXXIX. Singh Munish; Thawkar, Shankar; Singh, Rekha, L.K.K. (2023) ‘Deep-learning based system for effective and automatic blood vessel segmentation from Retinal fundus images’, Multimedia Tools and Applications, 83(2), pp. 6005–6049. 10.1007/s11042-023-15348-3.
XL. Sobisch, J. et al. (2026) ‘Toward Well-Connected Retina Segmentation: A Fully Differentiable Endpoint Connectivity Loss (DECL)’, IEEE Access, 14, pp. 63326–63341. 10.1109/ACCESS.2026.3686973.
XLI. Spaide James G.; Waheed, Nadia K., R.F.. F. (2015) ‘Image Artifacts in Optical Coherence Tomography Angiography’, Retina (Philadelphia, Pa.), 35(11), pp. 2163–2180. 10.1097/iae.0000000000000765.
XLII. Staal Michael D.; Niemeijer, Meindert; Viergever, Max A.; van Ginneken, B., J.A. (2004) ‘Ridge-based vessel segmentation in color images of the retina’, IEEE transactions on medical imaging, 23(4), pp. 501–509. 10.1109/tmi.2004.825627.
XLIII. Szegedy, C. et al. (2015) ‘Going deeper with convolutions’, in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–9. 10.1109/CVPR.2015.7298594.
XLIV. Topta? Davut, B.H. (2021) ‘Retinal blood vessel segmentation using pixel-based feature vector’, Biomedical Signal Processing and Control, 70(NA), pp. 103053-NA. 10.1016/j.bspc.2021.103053.
XLV. Uysal Gur Emre, E.G. (2020) ‘Computer-aided retinal vessel segmentation in retinal images: convolutional neural networks’, Multimedia Tools and Applications, 80(3), pp. 3505–3528. 10.1007/s11042-020-09372-w.
XLVI. Wan Gaofeng; Li, Renxing; Xiang, Yifan; Yin, Dechao; Yang, Minglei; Gong, Deren; Chen, Jiangang, Y.W. (2024) ‘Retinal Blood Vessels Segmentation With Improved <scp>SE?UNet</scp> Model’, International Journal of Imaging Systems and Technology, 34(4), p. NA-NA. 10.1002/ima.23145.
XLVII. Wang Xiaobo; Ma, Zhendi, J.L. (2025) ‘Multi-Scale Three-Path Network (MSTP-Net): A new architecture for retinal vessel segmentation’, Measurement, 250(NA), p. 117100. 10.1016/j.measurement.2025.117100.
XLVIII. Wu Yong; Song, Yang; Zhang, Yanning; Cai, Weidong, Y.X. (2020) ‘NFN?: A novel network followed network for retinal vessel segmentation.’, Neural networks?: the official journal of the International Neural Network Society, 126(NA), pp. 153–162. 10.1016/j.neunet.2020.02.018.
XLIX. Yaz?c?, Z.A. et al. (2026) ‘VEELA: A Clinically-Constrained Benchmark for Liver Vessel Segmentation in Computed Tomography Angiography’.
L. Zafar, A. et al. (2022) ‘A Comparison of Pooling Methods for Convolutional Neural Networks’, Applied Sciences, 12(17). 10.3390/app12178643.
LI. Zeiler, M.D. and Fergus, R. (2013) ‘Visualizing and Understanding Convolutional Networks’, CoRR, abs/1311.2.
LII. Zhu Xiang; Zhang, Xue-Dian; Jiang, Min-Shan, Y.-F.X. (2023) ‘CCS-UNet: a cross-channel spatial attention model for accurate retinal vessel segmentation.’, Biomedical optics express, 14(9), p. 4739. 10.1364/boe.495766.