Abstract
There are several novel applications of Deep Learning in Medical Imaging and especially in Ophthalmology in order to provide solutions to unmet clinical needs. The research presented in this paper concerns semantic segmentation of lesions regarding Diabetic Retinopathy. Most of the state-of-the-art papers nowadays use Convolutional Neural Networks, Fully Convolutional Networks, and UNETs, a modified version of Convolutional Neural Networks for segmentation tasks. The robustness of UNETs, in conjunction with transfer learning, has been the main strategy to tackle the limitations of the available public datasets. In this paper, the encoder of a UNET has been substituted by MobileNetV2, which constitutes a novel approach for segmenting Diabetic Retinopathy lesions. Results show that the proposed model, in hemorrhages and soft exudates lesions surpasses other similar attempts. In the proposed model, sensitivity reached 0.89 in hemorrhages and 0.97 in soft exudates. Another novelty of the paper is that the results are further analyzed at the lesion level, in contrast to the common pixel-level analysis met in the literature, something that favors a more intuitive evaluation of the model.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Centers for Disease Control and Prevention (CDC): What is diabetes? May 2021. https://www.cdc.gov/diabetes/basics/diabetes.html. Accessed 28 Nov 2021
Chen, L.C., Papandreou, G., Schroff, F., Adam, H.: Rethinking atrous convolution for semantic image segmentation (2017)
Chudzik, P., Majumdar, S., Calivá, F., Al-Diri, B., Hunter, A.: Exudate segmentation using fully convolutional neural networks and inception modules. In: Angelini, E.D., Landman, B.A. (eds.) Medical Imaging 2018: Image Processing. SPIE, March 2018. https://doi.org/10.1117/12.2293549
Eftekhari, N., Pourreza, H.R., Masoudi, M., Ghiasi-Shirazi, K., Saeedi, E.: Microaneurysm detection in fundus images using a two-step convolutional neural network. BioMedical Eng. On Line 18(67) (2019). https://doi.org/10.1186/s12938-019-0675-9
Furtado, P.: Segmentation of diabetic retinopathy lesions by deep learning: achievements and limitations. In: 7th International Conference on Bioimaging, pp. 95–101. SCITEPRESS - Science and Technology Publications, January 2020. https://doi.org/10.5220/0008881100950101
Furtado, P.: Using segmentation networks on diabetic retinopathy lesions: metrics, results and challenges. In: Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021). BIOIMAGING, vol. 2, pp. 128–135. SCITEPRESS - Science and Technology Publications (2021). https://doi.org/10.5220/0010208501280135
(IDRiD), I.D.R.I.D., October 2017. https://idrid.grand-challenge.org. Accessed 20 Aug 2021
ImageNet: March 2021. https://www.image-net.org/index.php. Accessed 29 Aug 2021
keras.io: adam. (2018). https://keras.io/api/optimizers/adam/. Accessed 27 Nov 2021
Khojasteh, P., Aliahmad, B., Kumar, D.K.: Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms. BMC Ophthalmol. 18(1) (2018). https://doi.org/10.1186/s12886-018-0954-4
Khojasteh, P., et al.: Exudate detection in fundus images using deeply-learnable features. Comput. Biol. Med. 104, 62–69 (2019). https://doi.org/10.1016/j.compbiomed.2018.10.031
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation (2015)
Perdomo, O., Arevalo, J., González, F.A.: Convolutional network to detect exudates in eye fundus images of diabetic subjects. In: Romero, E., Lepore, N., Brieva, J., Larrabide, I. (eds.) 12th International Symposium on Medical Information Processing and Analysis. SPIE, January 2017. https://doi.org/10.1117/12.2256939
Popli, A., Jindal, G., Pillai, G., Khan, H.R., Agarwal, M., Yadav, V.: Automated hard exudates segmentation in retinal images using patch based UNet, July 2018. https://github.com/apopli/diabetic-retinopathy/blob/master/segmentation-hard-exudates.pdf
Appan K., P., Sivaswamy, J.: Retinal image synthesis for CAD development. In: Campilho, A., Karray, F., ter Haar Romeny, B. (eds.) ICIAR 2018. LNCS, vol. 10882, pp. 613–621. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-93000-8_70
Si, Z., Fu, D., Liu, Y., Huang, Z.: Hard exudate segmentation in retinal image with attention mechanism. IET Image Process. 15(3), 587–597 (2020). https://doi.org/10.1049/ipr2.12007
Siddique, N., Paheding, S., Elkin, C.P., Devabhaktuni, V.: U-net and its variants for medical image segmentation: a review of theory and applications. IEEE Access 9, 82031–82057 (2021). https://doi.org/10.1109/access.2021.3086020
tensorflow.org: tf.keras.preprocessing.image.ImageDataGenerator, November 2021. https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator. Accessed 20 Nov 2021
Tiu, E.: Metrics to evaluate your semantic segmentation model, August 2019. https://towardsdatascience.com/metrics-to-evaluate-your-semantic-segmentation-model-6bcb99639aa2. Accessed 15 Nov 2021
Tsiknakis, N., et al.: Deep learning for diabetic retinopathy detection and classification based on fundus images: a review. Comput. Biol. Med. 135, 104599 (2021). https://doi.org/10.1016/j.compbiomed.2021.104599, https://www.sciencedirect.com/science/article/pii/S0010482521003930
Usman Akram, M., Khalid, S., Tariq, A., Khan, S.A., Azam, F.: Detection and classification of retinal lesions for grading of diabetic retinopathy. Comput. Biol. Med. 45, 161–171 (2014). https://doi.org/10.1016/j.compbiomed.2013.11.014, https://www.sciencedirect.com/science/article/pii/S0010482513003430
Wang, W., Hu, Y., Zou, T., Liu, H., Wang, J., Wang, X.: A new image classification approach via improved MobileNet models with local receptive field expansion in shallow layers. Comput. Intell. Neurosci. 2020, 1–10 (2020). https://doi.org/10.1155/2020/8817849
Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions, April 2016, version 3
Zheng, R., et al.: Detection of exudates in fundus photographs with imbalanced learning using conditional generative adversarial network. Biomed. Opt. Exp. 9(10), 4863–4878 (2018). https://doi.org/10.1364/boe.9.004863
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Theodoropoulos, D., Manikis, G.C., Marias, K., Papadourakis, G. (2022). Semantic Segmentation of Diabetic Retinopathy Lesions, Using a UNET with Pretrained Encoder. In: Iliadis, L., Jayne, C., Tefas, A., Pimenidis, E. (eds) Engineering Applications of Neural Networks. EANN 2022. Communications in Computer and Information Science, vol 1600. Springer, Cham. https://doi.org/10.1007/978-3-031-08223-8_30
Download citation
DOI: https://doi.org/10.1007/978-3-031-08223-8_30
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-08222-1
Online ISBN: 978-3-031-08223-8
eBook Packages: Computer ScienceComputer Science (R0)