A Patch - Based Analysis for Retinal Lesion Segmentation with Deep Neural Networks

  • A Mary DayanaEmail author
  • W. R. Sam Emmanuel
Conference paper
Part of the Lecture Notes on Data Engineering and Communications Technologies book series (LNDECT, volume 49)


Diabetic Retinopathy (DR) is one of the key symptoms of Diabetes Mellitus that is caused by the deterioration of blood capillaries which nourish the retina of the eye. The spread of the pathological lesions in the retina determine the severity of diabetic retinopathy. Therefore, it is obligatory to analyze the symptoms of DR at an early clinical stage as it prevents the progression of the disease and protects the vision. Deep learning algorithms play an important role in detecting multiple abnormalities from retinal fundus images and also highlights the areas of corresponding lesions with considerable accuracy. In the proposed method a deep convolutional neural network is designed for the segmentation of small lesions with patch-based analysis in retinal fundus images. A sliding window method was used to create image patches. The trained network analyzes the image patches and generates the probability map which in turn predicts the different types of lesions. The results obtained by the proposed work shows significantly better performance in accuracy and sensitivity when compared with other works on related tasks.


Diabetic Retinopathy Diabetes Mellitus Deep learning Convolutional Neural Network Sliding window 


  1. 1.
    IDF Diabetes Atlas (2017)Google Scholar
  2. 2.
    Roychowdury, S., Koozekanani, D.D., Parhi, K.K.: DREAM: Diabetic retinopathy analysis using machine learning. IEEE J. Biomed. Health Inform. 18(5), 1717–1728 (2014)CrossRefGoogle Scholar
  3. 3.
    Gao, Z., et al.: Diagnosis of diabetic retinopathy using deep neural networks. IEEE Access 7, 3360–3370 (2019)CrossRefGoogle Scholar
  4. 4.
    Lam, C., Yu, C., Huang, L., Rubin, D.: Retinal lesion detection with deep learning using image patches. Invest. Ophthalmol. Vis. Sci. 59(1), 590–596 (2018)CrossRefGoogle Scholar
  5. 5.
    Khojasteh, P., Aliahmad, B., Kumar, D.K.: Fundus images analysis using deep features for detection of exudates, hemorrhages, and microaneurysms. BMC Ophthalmol. 18(1), 1–13 (2018)CrossRefGoogle Scholar
  6. 6.
    Yang, Y., Li, T., Li, W., Wu, H., Fan, W.: Lesion detection and grading of diabetic retinopathy via two-stages deep convolutional neural networks. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 533–540 (2017)CrossRefGoogle Scholar
  7. 7.
    Chandore, V., Asati, S.: Automatic detection of diabetic retinopathy using deep convolutional neural network. Int. J. Adv. Res. Ideas Innov. Technol. 3, 633–641 (2017)Google Scholar
  8. 8.
    Tan, J.H., et al.: Automated segmentation of exudates, hemorrhages, microaneurysms using a single convolutional neural network. Inf. Sci. 420, 66–76 (2017)CrossRefGoogle Scholar
  9. 9.
    Gargeya, R., Leng, T.: Automated identification of diabetic retinopathy using deep learning. Ophthalmology 124(7), 962–969 (2017)CrossRefGoogle Scholar
  10. 10.
    Mansour, R.F.: Deep-learning-based automatic computer-aided diagnosis system for diabetic retinopathy. Biomed. Eng. Lett. 8, 41–47 (2018)CrossRefGoogle Scholar
  11. 11.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances In Neural Information Processing Systems (2012)Google Scholar
  12. 12.
    Guo, S., et al.: L-Seg: An end-to-end unified framework for multi-lesion segmentation of fundus images. Neurocomputing 349, 52–63 (2019)CrossRefGoogle Scholar
  13. 13.
    Gondal, W.M., Kohler, J.M., Grzeszick, R., Fink, G.A., Hirsch, M.: Weakly-supervised localization of diabetic retinopathy lesions in retinal fundus images. In: Proceedings of International Conference on Image Processing (ICIP), pp. 2069–2073 (2018)Google Scholar
  14. 14.
    Wan, S., Liang, Y., Zhang, Y.: Deep convolutional neural networks for diabetic retinopathy detection by image classification. Comput. Electr. Eng. 72, 274–282 (2018)CrossRefGoogle Scholar
  15. 15.
    Gulshan, V., Peng, L., Coram, M., et al.: Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. J. Am. Med. Assoc. 316(22), 2402–2410 (2016)CrossRefGoogle Scholar
  16. 16.
    Orlando, J.I., Prokofyeva, E., deFresno, M., Blaschko, M.B.: An ensemble deep learning-based approach for red lesion detection in fundus ımages. Comput. Methods Programs Biomedi. 153, 115–127 (2018)CrossRefGoogle Scholar
  17. 17.
    Son, J., et al.: Development and validation of deep learning models for screening multiple abnormal findings in retinal fundus ımages. Ophthalmology 127, 85–94 (2019)CrossRefGoogle Scholar
  18. 18.
    Badar, M., Shahzad, M., Fraz M.M.: Simultaneous Segmentation of Multiple Retinal Pathologies Using Fully Convolutional Deep Neural Network, pp. 313–324. Springer, Cham (2018)Google Scholar
  19. 19.
    Wang, X., Lu, Y., Wang, Y., Chen, W.: Diabetic retinopathy stage classification using convolutional neural networks. In: 2018 IEEE International Conference on Information Reuse and Integration (IRI), pp. 465–471 (2018)Google Scholar
  20. 20.
    Nitish, S., Geoffrey, H., Alex, K., Hya, S., Ruslan, S.: Dropout: A simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2018)MathSciNetzbMATHGoogle Scholar
  21. 21.
    Kauppi, T., et al.: The DIARETDB1 diabetic retinopathy database and evaluation protocol. J. Med. Imaging 3, 1–18 (2007)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of Computer ScienceNesamony Memorial Christian College, Affiliated to Manonmaniam Sundaranar UniversityTirunelveliIndia

Personalised recommendations