Modified U-Net architecture for semantic segmentation of diabetic retinopathy images

• Autorzy: Sambyal Nitigya , Saini Poonam , Syal Rupali , Gupta Varun

Identyfikatory

DOI

10.1016/j.bbe.2020.05.006

• Języki publikacji: EN

Abstrakty

EN

Segmentation of lesions from fundus images is an essential prerequisite for accurate severity assessment of diabetic retinopathy. Due to variation in morphologies, number and size of lesions, the manual grading process becomes extremely challenging and time-consuming. This necessitates the need of an automatic segmentation system that can precisely define the region of interest boundaries and assist ophthalmologists in speedy diagnosis along with diabetic retinopathy severity grading. The paper presents a modified U-Net architecture based on residual network and employs periodic shuffling with sub-pixel convolution initialized to convolution nearest neighbour resize. The proposed architecture has been trained and validated for microaneurysm and hard exudate segmentation on two publicly available datasets namely IDRiD and e-ophtha. For IDRiD dataset, the network obtains 99.88% accuracy, 99.85% sensitivity, 99.95% specificity and dice score of 0.9998 for both microaneurysm and exudate segmentation. Further, when trained on e-ophtha and validated on IDRiD dataset, the network shows 99.98% accuracy, 99.88% sensitivity, 99.89% specificity and dice score of 0.9998 for microaneurysm segmentation. For exudates segmen-tation, the model obtained 99.98% accuracy, 99.88% sensitivity, 99.89% specificity and dice score of 0.9999, when trained on e-ophtha and validated on IDRiD dataset. In comparison to existing literature, the proposed model provides state-of-the-art results for retinal lesion segmentation.

Słowa kluczowe

EN

fundus image; microaneurysm; hard exudates; semantic segmentation; modified U-Net; diabetic retinopathy

PL

obraz dna oka; wysięk twardy; segmentacja semantyczna; retinopatia cukrzycowa

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2020
• Tom: Vol. 40, no. 3
• Strony: 1094--1109
• Opis fizyczny: Bibliogr. 34 poz., rys., tab., wykr.

Twórcy

autor

Sambyal Nitigya

• Department of Computer Science & Engineering, Punjab Engineering College (Deemed to be University) , Chandigarh 160012, India

autor

Saini Poonam

• Department of Computer Science & Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India

autor

Syal Rupali

• Department of Computer Science & Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India

autor

Gupta Varun

• Department of Computer Science & Engineering, Chandigarh College of Engineering and Technology, Chandigarh, India

Bibliografia

• [1] Devaraj DRS, Prasanna Kumar SC. A survey on segmentation of exudates and microaneurysms for early detection of diabetic retinopathy. Mater Today Proc 2018;10845–50.
• [2] Joshi S, Karule PT. A review on exudates detection methods for diabetic retinopathy. Biomed Pharmacother 2017;97:1454–60.
• [3] Wan S, Liang Y, Zhang Y. Deep convolutional neural networks for diabetic retinopathy detection by image classification. Comput Electr Eng 2018;72:274–82.
• [4] Shanthi T, Sabeenian RS. Modified Alexnet architecture for classification of diabetic retinopathy images. Comput Electr Eng 2019;76:56–64.
• [5] Chudzik P, Al-diri B, Caliv F, Ometto G, Hunter A. Exudates segmentation using fully convolutional neural network and auxiliary codebook. 2018 40th Annu Int Conf IEEE Eng Biol Soc; 2018.
• [6] Qomariah D, Tjandrasa H. Exudate detection in retinal fundus images using combination of mathematical morphology and reny entropy thresholding. 2017 Int Conf Inf Commun Technol Syst; 2017.
• [7] Kaur J, Mittal D. Segmentation and measurement of exudates in fundus images of the retina for detection of retinal disease. J Biomed Eng Med Imaging 2015;2(1):27–38.
• [8] Karkuzhali S, Manimegalai D. Retinal haemorrhages segmentation using improved toboggan segmentation algorithm in diabetic retinopathy images. Biomed Res 2018;105–7.
• [9] Zubair M. Automated segmentation of hard exudates using dynamic thresholding to detect diabetic retinopathy in retinal photographs. Int Conf Innov Comput; 2016.
• [10] Long S, Huang X, Chen Z, Pardhan S, Zheng D. Automatic detection of hard exudates in color retinal images using dynamic threshold and SVM classification: algorithm development and evaluation. Biomed Res Int 2019;1–13.
• [11] Chudzik P, Majumdar S, Calivá F, Al-Diri B, Hunter A. Exudate segmentation using fully convolutional neural networks and inception modules. Proc SPIE Med Imaging 2018: Image Process 2018.
• [12] Saha O, Sathish R, Sheet D. Fully convolutional neural network for semantic segmentation of anatomical structure and pathologies in colour fundus images associated with diabetic retinopathy; 2019, arXiv:190203122v1[CsCV].
• [13] Lian S, Li L, Lian G, Xiao X, Luo Z, Li S. A global and local enhanced residual U-Net for accurate retinal vessel segmentation. IEEE/ACM Trans Comput Biol Bioinforma 2015;14:1–10.
• [14] Xu X, Wang R, Tan T, Xu F. An improved U-net architecture for simultaneous arteriole and venule segmentation in fundus image. Med Image Underst Anal 2018;1–10.
• [15] Xiancheng W, Wei L, Bingyi M, He J, Jiang Z, Xu W, et al. Retina blood vessel segmentation using a U-Net based convolutional neural network. Proc Comput Sci 2018.
• [16] Wang C, Zhao Z, Ren Q, Xu Y, Yu Y. Dense U-net based on patch-based learning for retinal vessel segmentation. Entropy 2019;21:1–15.
• [17] Jiang Y, Tan N, Peng T, Zhang H. Retinal vessels segmentation based on dilated multi-scale convolutional neural network. IEEE Access 2019;7:76342–5.
• [18] Li Q, Fan S, Chen C. An intelligent segmentation and diagnosis method for diabetic retinopathy based on improved U-Net network. J Med Syst 2019;43(9):1–9.
• [19] Arsalan M, Owais M, Mahmood T, Cho SW, Park KR. Aiding the diagnosis of diabetic and hypertensive retinopathy using artificial intelligence-based semantic segmentation. J Clin Med 2019;8:1–28.
• [20] Khakzar M, Pourghassem H. A retinal image authentication framework based on a graph-based representation algorithm in a two-stage matching structure. Biocybern Biomed Eng 2017;37:742–59.
• [21] Kaur J, Mittal D. A generalized method for the detection of vascular structure in pathological retinal images. Biocybern Biomed Eng 2017;37:184–200.
• [22] Jiang Z, Yepez J, An S, Ko S. Fast, accurate and robust retinal vessel segmentation system. Biocybern Biomed Eng 2017;37:412–21.
• [23] Gao X, Cai Y, Qiu C, Cui Y. Retinal blood vessel segmentation based on the Gaussian matched filter and U-net. 10th Int Congr Image Signal Process Biomed Eng Informatics 2017.
• [24] Challoob M, Gao Y. Retinal vessel segmentation using matched filter with joint relative entropy. Lect Notes Comput Sci 2017;10424:228–39.
• [25] Braovic´ M, Stipanicev D, Šeric´ L. Retinal blood vessel segmentation based on heuristic image analysis. Comput Sci Inf Syst 2019;16:227–45.
• [26] Navab N, Hornegger J, Wells WM, Frangi AF. U-Net: convolutional networks for biomedical image segmentation. Lect Notes Comput Sci 2015;9351:234–41.
• [27] Porwal P, Pachade S, Kamble R, Kokare M, Deshmukh G, Sahasrabuddhe V, et al. Indian diabetic retinopathy image dataset (IDRiD): a database for diabetic retinopathy screening research. Data 2018;3:1–8.
• [28] Decencière E, Cazuguel G, Zhang X, Thibault G, Klein J-C, Meyer F, et al. TeleOphta: machine learning and image processing methods for teleophthalmology. Innov Res Biomed Eng 2013;34:196–203.
• [29] Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation. Lect Notes Comput Sci 2015;9351:234–41.
• [30] He Kaiming, Zhang Xiangyu, Shaoqing Ren JS. Deep residual learning for image recognition. 2016 IEEE Conf Comput Vis Pattern Recognit. 2016. pp. 770–8.
• [31] Aitken A, Ledig C, Theis L, Caballero J, Wang Z, Shi W. Checkerboard artifact free sub-pixel convolution: a note on sub-pixel convolution, resize convolution and convolution resize; 2017, arXiv:170702937 [CsCV].
• [32] Shi W, Caballero J, Huszar F, Totz J, Aitken AP, Bishop R, et al. Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network. IEEE Comput Soc Conf Comput Vis Pattern Recogn; 2016.
• [33] Imani E, Pourreza H. A novel method for retinal exudate segmentation using signal seperation algorithm. Comput Methods Programs Biomed 2016;133:195–205.
• [34] Kaur J, Mittal D. A generalized method for the segmentation of exudates from pathological retinal fundus images. Biocybern Biomed Eng 2018;38:27–53.

Uwagi

PL

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).