Abstract
Long-term exposure to diabetes mellitus leads to the formation of diabetic retinopathy (DR), which can cause vision loss in working-age adults. Early stage diagnosis of DR is highly essential for preventing vision loss and preserving vision in people with diabetes. The motivation behind the severity grade classification of DR is to develop an automated system that can assist ophthalmologists and healthcare professionals in the diagnosis and management of DR. However, existing methods suffer from variability in image quality, similar structures of the normal and lesion regions, high dimensional features, variability in disease manifestations, small datasets, high training loss, model complexity, and overfitting, which leads to high misclassification errors in the severity grading system. Hence, there is a need to develop an automated system using improved deep learning techniques to provide a reliable and consistent grading of DR severity with high classification accuracy using fundus images. To solve these issues, we proposes a Deformable Ladder Bi attention U-shaped encoder-decoder network and Deep Adaptive Convolutional Neural Network (DLBUnet-DACNN) for accurate severity classification of DR. The DLBUnet performs lesion segmentation that can be divided into three parts: the encoder, the central processing module and the decoder. In the encoder part, deformable convolution is used instead of convolution to learn different shapes of the lesion by understanding the offset location. Afterwards, Ladder Atrous Spatial Pyramidal Pooling (LASPP) using variable dilation rates is introduced in the central processing module. LASPP enhance the tiny lesion features and variable dilation rates avoid gridding effects and can learn better global context information. Then the decoder part uses a bi-attention layer contains spatial and channel attention, which can learn contour and edges of the lesion accurately. Finally, the severity of DR is classified using a DACNN by extracting the discriminative features from the segmentation results. Experiments are conducted on the Messidor-2, Kaggle, and Messidor datasets. Our proposed method DLBUnet-DACNN achieves better results in terms of accuracy of 98.2, recall of 0.987, kappa coefficient of 0.993, precision of 0.98, F1-score of 0.981, Matthews Correlation Coefficient (MCC) of 0.93 and Classification Success Index (CSI) of 0.96 when compared to existing methods.
Graphical Abstract
Similar content being viewed by others
References
Gadekallu TR, Khare N, Bhattacharya S, Singh S, Maddikunta PKR, Srivastava G (2020) Deep neural networks to predict diabetic retinopathy. J Ambient Intell Humanized Comput 14(5):5407–5420
Bandello F, Toni D, Porta M, Varano M (2020) Diabetic retinopathy, diabetic macular edema, and cardiovascular risk: the importance of a long-term perspective and a multidisciplinary approach to optimal intravitreal therapy. Acta Diabetol 57(5):513–526
Rathmann W, Kostev K (2015) Fracture risk in patients with newly diagnosed type 2 diabetes: a retrospective database analysis in primary care. J Diabetes Complications 29(6):766–770
Sun Y, Roumelioti ME, Ganta K, Glew RH, Gibb J, Vigil D, Tzamaloukas AH (2020) Dialysis-associated hyperglycemia: manifestations and treatment. Int Urol Nephrol 52(3):505–517
Bhardwaj C, Jain S, Sood M (2021) Hierarchical severity grade classification of non-proliferative diabetic retinopathy. J Ambient Intell Humanized Comput 12(2):2649–2670
Zeng Y, Cao D, Yang D, Zhuang X, Hu Y, He M, Zhang L (2020) Retinal vasculature–function correlation in non-proliferative diabetic retinopathy. Doc Ophthalmol 140(2):129–138
Gayathri S, Gopi VP, Palanisamy P (2020) Automated classification of diabetic retinopathy through reliable feature selection. Phys Eng Sci Med 43(3):927–945
Gouder MJ (2020) Retinal testing and imaging. In: The Retina. Springer, Cham, pp 9-21
Behera MK, Chakravarty S (2020) Diabetic retinopathy image classification using support vector machine. In: 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA). IEEE, Gunupur, India, pp 1–4
Osman AH, Aljahdali HMA (2020) An effective of ensemble boosting learning method for breast cancer virtual screening using neural network model. IEEE Access 8:39165–39174
Balakrishnan N, Nisi K (2020) A deep analysis on optimization techniques for appropriate PID tuning to incline efficient artificial pancreas. Neural Comput Appl 32(12):7587–7596
Raja JB, Pandian SC (2020) PSO-FCM based data mining model to predict diabetic disease. Comput Methods Programs Biomed 196:105659
Asiri N, Hussain M, Al Adel F, Alzaidi N (2019) Deep learning based computer-aided diagnosis systems for diabetic retinopathy: a survey. Artif Intell Med 99:101701
Islam MR, Abdulrazak LF, Nahiduzzaman M, Goni MO, Anower MS, Ahsan M, Haider J, Kowalski M (2022) Applying supervised contrastive learning for the detection of diabetic retinopathy and its severity levels from fundus images. Comput Biol Med 146:105602
Berbar MA (2022) Features extraction using encoded local binary pattern for detection and grading diabetic retinopathy. Health Inf Sci Syst 10(1):1–3
Nahiduzzaman M, Islam MR, Islam SR, Goni MO, Anower MS, Kwak KS (2021) Hybrid CNN-SVD based prominent feature extraction and selection for grading diabetic retinopathy using extreme learning machine algorithm. IEEE Access 9:152261–152274
Yang Y, Shang F, Wu B, Yang D, Wang L, Xu Y, Zhang W, Zhang T (2021) Robust collaborative learning of patch-level and image-level annotations for diabetic retinopathy grading from fundus image. IEEE Trans Cybern 52:11407
Gangwar AK, Ravi V (2021) Diabetic retinopathy detection using transfer learning and deep learning. In: Evolution in Computational Intelligence. Springer, Singapore, pp 679–689
Sridhar S, PradeepKandhasamy J, Sinthuja M, Minish TS (2021) Diabetic retinopathy detection using convolutional nueral networks algorithm. Mater Today: Proc 14(9):1932. https://doi.org/10.3390/sym14091932
Kalyani G, Janakiramaiah B, Karuna A, Prasad LVN (2021) Diabetic retinopathy detection and classification using capsule networks. Complex Intell Syst 9:2651–2664
Badgujar RD, Deore PJ (2019) Hybrid nature inspired SMO-GBM classifier for exudate classification on fundus retinal images. IRBM 40(2):69–77
Dong B, Wang X, Qiang X, Du F, Gao L, Wu Q, Cao G, Dai C (2022) A multi-branch convolutional neural network for screening and staging of diabetic retinopathy based on wide-field optical coherence tomography angiography. IRBM 43(6):614–620
Han J, Wang Y, Gong H (2022) Fundus retinal vessels image segmentation method based on improved U-Net. IRBM 43(6):628–639
Novitasari DC, Fatmawati F, Hendradi R, Rohayani H, Nariswari R, Arnita A, Hadi MI, Saputra RA, Primadewi A (2022) Image fundus classification system for diabetic retinopathy stage detection using hybrid CNN-DELM. Big Data Cogn Comput 6(4):146
Barua PD, Chan WY, Dogan S, Baygin M, Tuncer T, Ciaccio EJ, Islam N, Cheong KH, Shahid ZS, Acharya UR (2021) Multilevel deep feature generation framework for automated detection of retinal abnormalities using OCT images. Entropy 23(12):1651
Kobat SG, Baygin N, Yusufoglu E, Baygin M, Barua PD, Dogan S, Yaman O, Celiker U, Yildirim H, Tan RS, Tuncer T (2022) Automated diabetic retinopathy detection using horizontal and vertical patch division-based pre-trained DenseNET with digital fundus images. Diagnostics 12(8):1975
Zulfahmi R, Noviyanti DS, Utami GR, Harison AN, Agung PS (2019) Improved image quality retinal fundus with contrast limited adaptive histogram equalization and filter variation. In2019 International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS) (pp. 49–54). IEEE
Abdelmaksoud E, El-Sappagh S, Barakat S, Abuhmed T, Elmogy M (2021) Automatic diabetic retinopathy grading system based on detecting multiple retinal lesions. IEEE Access 9:15939–15960
Li M, Lian F, Guo S (2021) Automatic pancreas segmentation using double adversarial networks with pyramidal pooling module. IEEE Access 9:140965–140974
Chen LC, Papandreou G, Schroff F, Adam H (2017) Rethinking atrous convolution for semantic image segmentation. arXiv:1706.05587. https://doi.org/10.48550/arXiv.1706.05587
Jie F, Nie Q, Li M, Yin M, Jin T (2021) Atrous spatial pyramid convolution for object detection with encoder-decoder. Neurocomputing 464:107–118
Gegundez-Arias ME, Marin-Santos D, Perez-Borrero I, Vasallo-Vazquez MJ (2021) A new deep learning method for blood vessel segmentation in retinal images based on convolutional kernels and modified U-Net model. Comput Methods Programs Biomed 205:106081
Kaggle Dataset (2015). https://www.kaggle.com/c/diabetic-retinopathy-detection/data
Messidor Dataset (2014). https://www.adcis.net/en/third-party/messidor/
Messidor 2 Dataset (2021). https://www.adcis.net/en/third-party/messidor2/
Bock S, Goppold J, Weiß M (2018) An improvement of the convergence proof of the ADAM-Optimizer. arXiv preprint arXiv:1804.10587
Author information
Authors and Affiliations
Contributions
All the authors have participated in writing the manuscript and have revised the final version. All authors read and approved the final manuscript. All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Binny Jeba Durai D and Jaya T. The first draft of the manuscript was written by Binny Jeba Durai D and all authors commented on previous versions of the manuscript. Conceptualization: Binny Jeba Durai D; Methodology: Binny Jeba Durai D; Formal analysis and investigation: Binny Jeba Durai D, Jaya T; Writing—original draft preparation: Binny Jeba Durai D; Writing—review and editing: Jaya T; Supervision: Jaya T.
Corresponding author
Ethics declarations
Ethical approval
This article does not contain any studies with human participants and/or animals performed by any of the authors.
Informed consent
There is no informed consent for this study.
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Durai, D.B.J., Jaya, T. Automatic severity grade classification of diabetic retinopathy using deformable ladder Bi attention U-net and deep adaptive CNN. Med Biol Eng Comput 61, 2091–2113 (2023). https://doi.org/10.1007/s11517-023-02860-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-023-02860-9