For finding the ophthalmic diseases that can be observed from retinal vessels, fundus images play a key role and provide an effective reference for professional medical personnel. However, manual vessel segmentation has a large workload, which is time-consuming and laborious. Therefore, developing an automatic and intelligent segmentation method is of great benefit to relevant personnel. In this study, the attention mechanism and RU-Net structure are integrated into the generator of generative adversarial networks (GANs), forming a new structure—Retina-GAN. At the same time, automatic color equalization (ACE) is selected in the preprocessing of fundus images to improve image contrast and make blood vessels clearer. To validate the proposed approach, we compared the Retina-GAN with some other models on DRIVE datasets. Accuracy, sensitivity, and specificity are measured for comparative analysis. The experiment shows that Retina-GAN has better performance than other models.
ReferencesKhan MI, Shaikh H, Mansuri AMA review of retinal vessel segmentation techniques and algorithms20112511401144
Khan MI, Shaikh H, Mansuri AM. A review of retinal vessel segmentation techniques and algorithms. International Journal of Computer Technology & Applications, 2011, 2(5): 1140–1144.
Chaudhuri S, Chatterjee S, Katz N, et alDetection of blood vessels in retinal images using two-dimensional matched filters19898326326910.1109/42.34715
Chaudhuri S, Chatterjee S, Katz N, etal. Detection of blood vessels in retinal images using two-dimensional matched filters. IEEE Transactions on Medical Imaging, 1989, 8(3): 263–269.
Wang SL, Yin YL, Cao GC, et alHierarchical retinal blood vessel segmentation based on feature and ensemble learning201514970871710.1016/j.neucom.2014.07.059
Wang SL, Yin YL, Cao GC, etal. Hierarchical retinal blood vessel segmentation based on feature and ensemble learning. Neurocomputing, 2015, 149: 708–717.
et al. DeepVessel: Retinal vessel segmentation via deep learning and conditional random field. Proceedings of the 19th International Conference on Medical Image Computing and Computer-assisted Intervention. Athens: Springer, 2016. 132–139.]]>
Rammy SA, Abbas W, Hassan NU, et alCPGAN: Conditional patch-based generative adversarial network for retinal vessel segmentation20201461081109010.1049/iet-ipr.2019.1007
Rammy SA, Abbas W, Hassan NU, etal. CPGAN: Conditional patch-based generative adversarial network for retinal vesselsegmentation. IET Image Processing, 2020, 14(6): 1081–1090.
Alom MZ, Yakopcic C, Hasan M, et alRecurrent residual U-Net for medical image segmentation201961014006
Alom MZ, Yakopcic C, Hasan M, etal. Recurrent residual U-Net for medical image segmentation. Journal of Medical Imaging, 2019, 6(1): 014006.
et al. Generative adversarial net. Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal: MIT Press, 2014. 2672–2680.]]>
et al. Image-to-image translation with conditional adversarial networks. Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017. 5967–5976.]]>
Rizzi A, Gatta C, Marini DFrom retinex to automatic color equalization: Issues in developing a new algorithm for unsupervised color equalization2004131758410.1117/1.1635366
Rizzi A, Gatta C, Marini D. From retinex to automatic color equalization: Issues in developing a new algorithm for unsupervised color equalization. Journal of Electronic Imaging, 2004, 13(1): 75–84.
Getreuer PAutomatic color enhancement (ACE) and its fast implementation2012226627710.5201/ipol.2012.g-ace
Getreuer P. Automatic color enhancement (ACE) and its fast implementation. Image Processing on Line, 2012, 2: 266–277.
et al. Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc., 2017. 6000–6010.]]>