Abstract
In the diagnosis of diabetic retinopathy and macula edema, the extraction of exudates is a crucial task, particularly in the presence of optic disc and cotton wools which have similar intensity. To add to this some of the reflectance around the vessels originating from the optic disc also needs to be eliminated. In this paper, supervised method with deep learning neural network is proposed for exudate classification. The network is trained with large number of sub-images, around 48,000 for each image, which are fed as inputs to the network. The original image is pre-processed before the patches are extracted. The database e-Ophtha is used. It has 47 images with exudates and ground truths. The performance of the system is good with an accuracy of 98.67%, area under the curve 97.29, sensitivity of 72.26% and specificity of 98.76%. Analysis of the system with respect to depth of network and patch size is also performed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Wild, S., Roglic, G., Green, A., Sicree, R., King, H.: Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004(27), 1047–1053 (2004)
Lundquist, M.B., Sharma, N., Kewalramani, K., Lundquist, M.B., Sharma, N., Kewalramani, K.: Patient perceptions of eye disease and treatment in Bihar India. J. Clin. Exp. Ophthalmol. 3, 213 (2012)
Goh, J.K.H., Sim, S.S., Tan, G.S.W.: Retinal imaging techniques for diabetic retinopathy screening. J. Diabetes Sci. Technol. 10, 282–294 (2016)
Decencière, E., Cazuguel, G., Zhang, X., Thibault, G., Klein, J.C., Meyer, F., Marcotegui, B., Quellec, G., Lamard, M., Danno, R., Elie, D., Massin, P., Viktor, Z., Erginary, A., Laÿ, B., Chabouis, A.: TeleOphta: machine learning and image processing methods for teleophthalmology. Innov. Res. BioMed. Eng. 34(2), 196–203 (2013)
Sinthanayothin, C., Kongbunkiat, V., Phoojaruenchanachai, S., Singalavanija A.: Automated screening system for diabetic retinopathy. In: Proceedings of the third International Symposium on Image and Signal Processing and Analysis, pp. 915–920 (2003)
Sopharak, A., Uyyanonvara, B., Barman, S., Williamson T.H.: Automatic detection of diabetic retinopathy exudates from non-dilated retinal images using mathematical morphology methods. J. Comput. Med. Imaging Graph. 32, 720–727 (2008)
Franklin, S.W., Rajan. S.E.: Diagnosis of diabetic retinopathy by employing image processing technique to detect exudates in retinal images. IET Image Proc. 8, 601–609 (2014)
Jaya, T., Dheeba, J., Albert Singh, N.: Detection of hard exudates in colour fundus images using fuzzy support vector machine-based expert system. J Digital Imaging 28, 761–768 (2015). Springer
Agurto, C., Murray, V., Yu, H., Wigdahl, J., Pattichis, M., Nemeth, S., Barriga, E.S., Soliz. P.: A multiscale optimization approach to detect exudates in the macula. IEEE J. Biomed. Health. Inf. 18(4),1328–1336 (2014)
Sandur, P., Naveena, C., Aradhya, V.N.M., Nagasundara, K.B.: Segmentation of brain tumor tissues in HGG and LGG MR images using 3D U-net convolutional neural network. Int. J. Nat. Comput. Res. 7(2), April–June (2018)
Liskowski, P., Krawiec, K.: Segmenting retinal blood vessels with deep neural networks. IEEE Trans. Med. Imaging 35(11), 2369–2380 (2016)
Ngo, L., Han, J.H.: Multi-level deep neural network for efficient segmentation of blood vessels in fundus images. Electron. Lett. 53(16), 1096–1098 (2017)
Cortinovis, D.: Retinal blood vessel segmentation with a convolutional neural network (U-net). https://github.com/orobix/retina-unet. Accessed 15 Sept 2018
Tan, J.H., Acharya, U.R., Bhandary, S.V., Chua, K.C., Sivaprasad, S.: Segmentation of optic disc, fovea and retinal vasculature using a single convolutional neural network. J. Comput. Sci. 20, 70–79 (2017)
Tan, J.H., Fujita, H., Sivaprasad, S., Bhandary, S.V., Rao, A.K., Chua, K.C., Acharya, U.R.: Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network. J. Inform. Sci. 420, 66–76 (2017)
Chen, T., Lin, L., Liu, L., Luo, X., Li, X.: DISC: deep image saliency computing via progressive representation learning. IEEE Trans. Neural Netw. Learn. Syst. 27(6), 1135–1149 (2016)
Quellec, G., Charrière, K., Boudi, Y., Cochener, B., Lamard, M.: Deep image mining for diabetic retinopathy screening. J. Med. Image Anal. 39, 178–193 (2017)
Harangi, B., Lazar, I., Hajdu, A.: Automatic exudate detection using active contour model and region wise classification. In: Engineering in Medicine and Biology Society (EMBC), Annual International Conference of the IEEE, pp. 5951–5954 (2012)
Prentašic, P., Loncǎric, S.: Detection of exudates in fundus photographs using deep neural networks and anatomical landmark detection fusion. Comput. Methods Programs Biomed. 137, 281–292 (2016)
Welfer, D., Scharcanski, J., Marinho. D.R.: A coarse-to-fine strategy for automatically detecting exudates in color eye fundus images. Comput. Med. Imaging Graph. 34, 228–235 (2010)
Mo, J., Zhang, L., Feng, Y.: Exudate-based diabetic macular edema recognition in retinal images using cascaded deep residual networks. J. Neurocomputing 290, 161–171 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Prabhu, N., Bhoir, D., Rao, U. (2020). Performance Analysis of Convolutional Neural Networks for Exudate Detection in Fundus Images. In: Bhateja, V., Satapathy, S., Zhang, YD., Aradhya, V. (eds) Intelligent Computing and Communication. ICICC 2019. Advances in Intelligent Systems and Computing, vol 1034. Springer, Singapore. https://doi.org/10.1007/978-981-15-1084-7_36
Download citation
DOI: https://doi.org/10.1007/978-981-15-1084-7_36
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-1083-0
Online ISBN: 978-981-15-1084-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)