Abstract
Segmentation of retinal structures, namely optic disc, vessel, demarcation line, and ridge, is essential for describing the characteristics of Retinopathy of Prematurity (ROP). Computerized systems are being developed for automatic segmentation in fundus images to assist the medical experts and bring consistency in the diagnosis. There are multiple challenges in the segmentation task of premature infants’ fundus images. The annotation and ground truth preparation required for the segmentation is complex, challenging, and expensive. Further, ROP datasets are not available publicly, and hence carrying out such a task needs a primary dataset and significant assistance from the domain expert. To address this gap, two primary datasets named HVDROPDB-BV and HVDROPDB-RIDGE were developed. The datasets consist of images captured by two different imaging systems, having different sizes, resolutions, and illumination. This made the trained models generic and robust to data variability and heterogeneity. We propose the modified U-Net architectures by incorporating squeeze and excitation (SE) blocks and attention gates (AG) to segment the demarcation line/ridge and vessel from these datasets. These modifications were tested and validated by ROP experts. The performance of all the three networks (U-Net, AG U-Net, and SE U-Net) was promising, with a variation of 1 to 6% in the dice coefficient for the HVDROPDB datasets. The area under the curve (AUC) obtained for all three networks was above 0.94, indicating them as excellent models. AG U-Net outperformed the other two networks, with a sensitivity of 96% and specificity of 89% for stage detection via new test images.
Similar content being viewed by others
References
Agrawal R, Agrawal M, Kulkarni S, Kotecha K, Walambe R (2021) Quantitative analysis of research on artificial intelligence in retinopathy of prematurity. Libr Philos Pract:1–29
Agrawal R, Kulkarni S, Walambe R, Kotecha K (2021) Assistive framework for automatic detection of all the zones in retinopathy of prematurity using deep learning. J Digit Imaging 34:1–16. https://doi.org/10.1007/s10278-021-00477-8
Ahmed, M. (2020). Medical image segmentation using attention-based deep neural networks.
Brown JM, Campbell JP, Beers A, Chang K, Ostmo S, Chan RP, Chiang MF (2018) Automated diagnosis of plus disease in retinopathy of prematurity using deep convolutional neural networks. JAMA Ophthalmol 136(7):803–810
Chen, C., Chuah, J. H., Raza, A., & Wang, Y. (2021). Retinal vessel segmentation using deep learning: a review. IEEE Access, Retinal Vessel Segmentation Using Deep Learning: A Review.
Chen C, Qin C, Qiu H, Tarroni G, Duan J, Bai W, Rueckert D (2020) Deep learning for cardiac image segmentation: a review. Front Cardiovasc Med 7:25
Chen, G., Zhao, J., Zhang, R., Wang, T., Zhang, G., & Lei, B. (2019, October). Automated stage analysis of retinopathy of prematurity using joint segmentation and multi-instance learning. In international workshop on ophthalmic medical image analysis (pp. 173-181). Springer, Cham.
Chen, L., Zhang, H., Xiao, J., Nie, L., Shao, J., Liu, W., & Chua, T. S. (2017). Sca-cnn: spatial and channel-wise attention in convolutional networks for image captioning. In proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5659-5667), arXiv:1611.05594
Ding, A., Chen, Q., Cao, Y., & Liu, B. (2020). Retinopathy of prematurity stage diagnosis using object segmentation and convolutional neural networks arXiv preprint arXiv: 2004.01582.
Du G, Cao X, Liang J, Chen X, Zhan Y (2020) Medical image segmentation based on u-net: a review. J Imaging Sci Technol 64(2):20508–20501
Gojić, G., Petrović, V., Turović, R., Dragan, D., Oros, A., Gajić, D., & Horvat, N. (2020). Deep learning methods for retinal blood vessel segmentation: evaluation on images with retinopathy of prematurity. In2020 IEEE 18th international symposium on intelligent systems and informatics (SISY)(pp. 131-136). IEEE.
Guan, Q., Huang, Y., Zhong, Z., Zheng, Z., Zheng, L., Yang, Y., 2018. Diagnose like a radiologist: attention guided convolutional neural network for thorax disease classification. arXiv:1801.09927.
Harouni, M., &Baghmaleki, H. Y. (2018). Color image segmentation metrics. Encyclopedia of image processing, 95, arXiv:2010.09907.
Hu J, Chen Y, Zhong J, Ju R, Yi Z (2018) Automated analysis for retinopathy of prematurity by deep neural networks. IEEE Trans Med Imaging 38(1):269–279
Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. InProceedings of the IEEE conference on computer vision and pattern recognition (pp. 7132-7141), https://doi.org/10.1109/CVPR.2018.00745.
Huang YP, Vadloori S, Chu HC, Kang EYC, Wu WC, Kusaka S, Fukushima Y (2020) Deep learning models for automated diagnosis of retinopathy of prematurity in preterm infants. Electronics 9(9):1444
Imran A, Li J, Pei Y, Yang JJ, Wang Q (2019) Comparative analysis of vessel segmentation techniques in retinal images. IEEE Access 7:114862–114887
International Committee for the Classification of Retinopathy of Prematurity (2005) The international classification of Retinopathy of prematurity revisited. Arch Ophthalmol (Chicago, Ill.: 1960) 123(7):991. https://doi.org/10.1001/archopht.123.7.991
Jetley, S., Lord, N. A., Lee, N., & Torr, P. H. (2018). Learn to pay attention. arXiv preprint arXiv:1804.02391.
Khanh TLB, Dao DP, Ho NH, Yang HJ, Baek ET, Lee G, Yoo SB (2020) Enhancing U-net with spatial-channel attention gate for abnormal tissue segmentation in medical imaging. Appl Sci 10(17):5729
Lei B, Zeng X, Huang S, Zhang R, Chen G, Zhao J, Wang T, Wang J, Zhang G (2021) Automated detection of retinopathy of prematurity by deep attention network. Multimed Tools Appl 80:36341–36360. https://doi.org/10.1007/s11042-021-11208-0
Milletari, F. (2018). Hough voting strategies for segmentation, detection and tracking (Doctoral dissertation, Technische Universität München).
Mittal K, Rajam VMA (2020) Computerized retinal image analysis-a survey. Multimed Tools Appl 79:22389–22421
Molinari A, Weaver D, Jalali S (2017) Classifying retinopathy of prematurity. Commu Eye Health 30(99):55–56
Mulay S, Ram K, Sivaprakasam M, Vinekar A (2019) Early detection of retinopathy of prematurity stage using deep learning approach. In medical imaging 2019: computer-aided diagnosis (Vol. 10950, p. 109502Z). Int Soc Optics Photonics. https://doi.org/10.1117/12.2512719
Murki S, Kadam S (2018) Role of a neonatal team including nurses in prevention of ROP. Comm Eye Health 31(101):S11–S15
Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M., Misawa, K., &Glocker, B. (2018). Attention u-net: learning where to look for the pancreas. arXiv preprint arXiv:1804.03999.
Razzak, M. I., Naz, S., &Zaib, A. (2018). Deep learning for medical image processing: overview, challenges and the future. In classification in BioApps (pp. 323–350). Springer, Cham, https://doi.org/10.1007/978-3-319-65981-7_12.
Ronneberger O., Fischer P., Brox T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab N., Hornegger J., Wells W., Frangi A. (eds) Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. MICCAI 2015. Lecture notes in computer science, vol 9351. Springer, Cham https://doi.org/10.1007/978-3-319-24574-4_28
Roy, A. G., Navab, N., &Wachinger, C. (2018). Concurrent spatial and channel' squeeze & excitation in fully convolutional networks. In international conference on medical image computing and computer-assisted intervention (pp. 421-429). Springer, Cham, arXiv:1803.02579.
Rundo L, Han C, Nagano Y, Zhang J, Hataya R, Militello C, Cazzaniga P (2019) USE-net: incorporating squeeze-and-excitation blocks into U-net for prostate zonal segmentation of multi-institutional MRI datasets. Neurocomputing 365:31–43
Schlemper J, Oktay O, Schaap M, Heinrich M, Kainz B, Glocker B, Rueckert D (2019) Attention gated networks: learning to leverage salient regions in medical images. Med Image Anal 53:197–207
Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6(1):1–48
Siddique, N., Sidike, P., Elkin, C., &Devabhaktuni, V. (2020). U-net and its variants for medical image segmentation: theory and applications. arXiv preprint arXiv:2011.01118, https://doi.org/10.1109/ACCESS.2021.3086020
Singh, N., Bansal, D., & Nagpal, D. (2020). Deep learning-based retinal vessel segmentation: a review. Adv. Math., Sci. J, 9(6), 3827-3837, DOI: https://doi.org/10.37418/amsj.9.6.62
Soomro TA, Afifi AJ, Zheng L, Soomro S, Gao J, Hellwich O, Paul M (2019) Deep learning models for retinal blood vessels segmentation: a review. IEEE Access 7:71696–71717
Tan Z, Simkin S, Lai C, Dai S (2019) A deep learning algorithm for automated diagnosis of retinopathy of prematurity plus disease. Trans Vision SciTechnol 8(6):23–23
Taylor S, Brown JM, Gupta K, Campbell JP, Ostmo S, Chan RP, Kalpathy-Cramer J (2019) Monitoring disease progression with a quantitative severity scale for retinopathy of prematurity using deep learning. JAMA Ophthalmol 137(9):1022–1028
Tiwari, S. S., Dholaria, A., Pandey, R., Nigam, G., Agrawal, R., Walambe, R., & Kotecha, K. (2020). Deep learning-based framework for retinal vasculature segmentation. In congress on intelligent systems (pp. 275-290). Springer, Singapore.
Tong Y, Lu W, Deng QQ, Chen C, Shen Y (2020) Automated identification of retinopathy of prematurity by image-based deep learning. Eye Vision 7(1):1–12
Uysal E, Güraksin GE (2021) Computer-aided retinal vessel segmentation in retinal images: convolutional neural networks. Multimed Tools Appl 80(3):3505–3528
Vijayalakshmi C, Sakthivel P, Vinekar A (2020) Automated detection and classification of telemedical retinopathy of prematurity images. Telemed e-Health 26(3):354–358. https://doi.org/10.1089/tmj.2019.0004
Wang, J., Lv, P., Wang, H., & Shi, C. (2021). SAR-U-net: squeeze-and-excitation block and atrous spatial pyramid pooling based residual U-net for automatic liver CT segmentation. arXiv preprint arXiv:2103.06419.
Yu Y, Zhu H (2021) Retinal vessel segmentation with constrained-based nonnegative matrix factorization and 3D modified attention U-net. EURASIP J Image Video Proces 2021(1):1–21
Acknowledgments
We thank Dr. Nilesh Giri, Dr. Pravin Hankare, and Dr. Anita Gaikwad from H. V. Desai Eye Hospital, Pune, for providing annotated images for research. We acknowledge H. V. Desai Eye Hospital staff’s help in providing daily fundus images to extend our current work. We also thank Mr. Anup Agrawal for helping in ground truth preparation using Adobe Photoshop.
Availability of data and material
PBMA’s H. V. Desai Eye Hospital, Pune.
Code availability
Available with authors—custom code.
Author information
Authors and Affiliations
Contributions
Ranjana Agrawal: Main author, creativity, programming, and system design, main conceptual work. Dr. Sucheta Kulkarni: Domain expert, guidance about the disease-specific concepts, dataset collection, and validation of results. Rahee Walambe: System design, conceptualization and ideation, and paper writing and review. Col. Madan Deshpande: Critical review of the manuscript. Ketan Kotecha: Conceptualization and preliminary assessment of work, paper review, senior author.
Corresponding author
Ethics declarations
There is nothing to declare by the authors.
Research involving human participants or animals
This study does not contain any studies with human participants or animals performed by any author.
Informed consent
Images obtained from preterm babies enrolled in the hospital’s screening program were used anonymously (without disclosing identity). As a protocol, written informed consent regarding the use of data for quality assurance and research purposes is obtained from preterm babies’ parents before screening for ROP.
Conflicts of interest/competing interests
The authors declare that they have no conflict of interest.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Agrawal, R., Kulkarni, S., Walambe, R. et al. Deep dive in retinal fundus image segmentation using deep learning for retinopathy of prematurity. Multimed Tools Appl 81, 11441–11460 (2022). https://doi.org/10.1007/s11042-022-12396-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-12396-z