Explainable AI Based Deep Ensemble Convolutional Learning for Multi-Categorical Ocular Disease Prediction

Abu Kowshir Bitto; Rezwana Karim; Mst Halema Begum; Md Fokrul Islam Khan Khan; Dr. Md. Maruf Hassan; Prof. Dr. Abdul kadar Muhammad Masum

doi:10.4108/airo.9234

Authors

Abu Kowshir Bitto Daffodil International University
Rezwana Karim Daffodil International University
Mst Halema Begum International American University
Md Fokrul Islam Khan Khan International American University
Dr. Md. Maruf Hassan Southeast University
Prof. Dr. Abdul kadar Muhammad Masum Southeast University

DOI:

https://doi.org/10.4108/airo.9234

Keywords:

Eye Disease, Deep Ensemble Learning , Transfer Learning , Explainable AI, Ocular Disease

Abstract

Diseases of the eye such as diabetic retinopathy, glaucoma, and cataract remain among the leading causes of blindness and vision impairment worldwide. Diagnosis in its early stages followed by early treatment is crucial to preventing permanent loss of vision. Recent advances in Artificial Intelligence (AI), particularly Transfer Learning and Explainable AI (XAI), have proven highly promising in automating the identification of retinal pathologies from medical images. In this paper, we propose an ensemble deep learning approach that integrates four pre-trained convolutional neural networks, i.e., VGG16, MobileNet, DenseNet, and InceptionV3, to classify retinal images into four categories: diabetic retinopathy, glaucoma, cataracts, and normal. The ensemble method leverages the power of multiple models to improve classification accuracy. Additionally, Explainable AI techniques are applied to make the model more interpretable, with visual explanations and insights into AI system decision-making and thereby establishing clinical trust and reliability. The system is evaluated on a new benchmarked eye disease dataset used from Hugging Face, and the results in terms of accuracy and model transparency are encouraging. This research contributes towards developing reliable, explainable, and efficient AI-driven diagnostic systems to assist healthcare professionals in the early detection and management of eye diseases

Downloads

Download data is not yet available.

References

[1] Bhadra AA, Jain M, Shidnal S. Automated detection of eye diseases. In2016 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) 2016 Mar 23 (pp. 1341-1345). IEEE.

[2] Jain L, Murthy HS, Patel C, Bansal D. Retinal eye disease detection using deep learning. In2018 Fourteenth International Conference on Information Processing (ICINPRO) 2018 Dec 21 (pp. 1-6). IEEE.

[3] Bock R, Meier J, Nyúl LG, Hornegger J, Michelson G. Glaucoma risk index: automated glaucoma detection from color fundus images. Medical image analysis. 2010 Jun 1;14(3):471-81.

[4] Sarki R, Ahmed K, Wang H, Zhang Y. Automatic detection of diabetic eye disease through deep learning using fundus images: a survey. IEEE access. 2020 Aug 10;8:151133-49.

[5] N. E. institute, “Facts About Diabetic Eye Disease,” https://nei.nih.gov/health/diabetic/retinopathy, 2015, [Online; Last Reviewed: September-2015].

[6] Prasad K, Sajith PS, Neema M, Madhu L, Priya PN. Multiple eye disease detection using Deep Neural Network. InTENCON 2019-2019 IEEE Region 10 Conference (TENCON) 2019 Oct 17 (pp. 2148-2153). IEEE.

[7] Abbas Q. Glaucoma-deep: detection of glaucoma eye disease on retinal fundus images using deep learning. International Journal of Advanced Computer Science and Applications. 2017;8(6).

[8] Dervisevic E, Pavljasevic S, Dervisevic A, Kasumovic SS. Challenges in early glaucoma detection. Medical Archives. 2016 May 31;70(3):203.

[9] Cvekl A, Vijg J. Aging of the eye: Lessons from cataracts and age-related macular degeneration. Ageing Research Reviews. 2024 Jul 6:102407.

[10] Bloemendal H, de Jong W, Jaenicke R, Lubsen NH, Slingsby C, Tardieu A. Ageing and vision: structure, stability and function of lens crystallins. Progress in biophysics and molecular biology. 2004 Nov 1;86(3):407-85.

[11] Roskamp KW, Paulson CN, Brubaker WD, Martin RW. Function and aggregation in structural eye lens crystallins. Accounts of chemical research. 2020 Apr 9;53(4):863-74.

[12] Lam D, Rao SK, Ratra V, Liu Y, Mitchell P, King J, Tassignon MJ, Jonas J, Pang CP, Chang DF. Cataract. Nature reviews Disease primers. 2015 Jun 11;1(1):1-5.

[13] Mrugacz M, Pony-Uram M, Bryl A, Zorena K. Current approach to the pathogenesis of diabetic cataracts. International Journal of Molecular Sciences. 2023 Mar 28;24(7):6317.

[14] Moraru AD, Costin D, Moraru RL, Branisteanu DC. Artificial intelligence and deep learning in ophthalmology-present and future. Experimental and therapeutic medicine. 2020 Oct;20(4):3469-73.

[15] Khan NC, Perera C, Dow ER, Chen KM, Mahajan VB, Mruthyunjaya P, Do DV, Leng T, Myung D. Predicting systemic health features from retinal fundus images using transfer-learning-based artificial intelligence models. Diagnostics. 2022 Jul 14;12(7):1714.

[16] Han D, Liu Q, Fan W. A new image classification method using CNN transfer learning and web data augmentation. Expert systems with applications. 2018 Apr 1;95:43-56.

[17] Morid MA, Borjali A, Del Fiol G. A scoping review of transfer learning research on medical image analysis using ImageNet. Computers in biology and medicine. 2021 Jan 1;128:104115.

[18] ul Hassan M, Al-Awady AA, Ahmed N, Saeed M, Alqahtani J, Alahmari AM, Javed MW. A transfer learning enabled approach for ocular disease detection and classification. Health Information Science and Systems. 2024 Jun 11;12(1):36.

[19] Chayan TI, Islam A, Rahman E, Reza MT, Apon TS, Alam MG. Explainable AI based glaucoma detection using transfer learning and LIME. In2022 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE) 2022 Dec 18 (pp. 1-6). IEEE.

[20] Cen LP, Ji J, Lin JW, Ju ST, Lin HJ, Li TP, Wang Y, Yang JF, Liu YF, Tan S, Tan L. Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks. Nature communications. 2021 Aug 10;12(1):4828.

[21] Kumar A, Dhanalakshmi R. EYE-YOLO: a multi-spatial pyramid pooling and Focal-EIOU loss inspired tiny YOLOv7 for fundus eye disease detection. International Journal of Intelligent Computing and Cybernetics. 2024 Jul 17;17(3):503-22.

[22] Shibata N, Tanito M, Mitsuhashi K, Fujino Y, Matsuura M, Murata H, Asaoka R. Development of a deep residual learning algorithm to screen for glaucoma from fundus photography. Scientific reports. 2018 Oct 2;8(1):14665.

[23] Aurangzeb K, Alharthi RS, Haider SI, Alhussein M. Systematic development of AI-enabled diagnostic systems for glaucoma and diabetic retinopathy. IEEE Access. 2023 Sep 20;11:105069-81.

[24] Sarhan MH, Nasseri MA, Zapp D, Maier M, Lohmann CP, Navab N, Eslami A. Machine learning techniques for ophthalmic data processing: a review. IEEE Journal of Biomedical and Health Informatics. 2020 Jul 28;24(12):3338-50.

[25] Alam M, Le D, Lim JI, Chan RV, Yao X. Supervised machine learning based multi-task artificial intelligence classification of retinopathies. Journal of clinical medicine. 2019 Jun 18;8(6):872.

[26] AlBalawi T, Aldajani MB, Abbas Q, Daadaa Y. IoT-Opthom-CAD: IoT-Enabled Classification System of Multiclass Retinal Eye Diseases Using Dynamic Swin Transformers and Explainable Artificial Intelligence. International Journal of Advanced Computer Science & Applications. 2024 Jul 1;15(7).

[27] Madduri VK, Rao BS. Detection and diagnosis of diabetic eye diseases using two phase transfer learning approach. PeerJ Computer Science. 2024 Sep 19;10:e2135.

[28] Mahmood SS, Chaabouni S, Fakhfakh A. Improving Automated Detection of Cataract Disease through Transfer Learning using ResNet50. Engineering, Technology & Applied Science Research. 2024 Oct 9;14(5):17541-7.

[29] Guo C, Yu M, Li J. Prediction of different eye diseases based on fundus photography via deep transfer learning. Journal of Clinical Medicine. 2021 Nov 23;10(23):5481.

[30] Aranha, G. D., Fernandes, R. A., & Morales, P. H. (2023). Deep transfer learning strategy to diagnose eye-related conditions and diseases: an approach based on low-quality fundus images. IEEE Access, 11, 37403-37411.

[31] Sanamdikar ST, Shelke MV, Rothe JP. Enhanced Classification of Diabetic Retinopathy via Vessel Segmentation: A Deep Ensemble Learning Approach. Journal homepage: http://iieta. org/journals/isi. 2023 Oct 1;28(5):1377-86.

[32] Desiani A, Primartha R, Hanum H, Dewi SR, Suprihatin B, Al-Filambany MG, Suedarmin M. Weighted Voting Ensemble Learning of CNN Architectures for Diabetic Retinopathy Classification. Jurnal Infotel. 2024 Feb 19;16(1):136-55.

[33] 1. Kumar, S. N. (N.D.). Nks9/Nks_Eye_Disease_Classification · Datasets At Hugging Face. Nks9/Nks_Eye_Disease_Classification· Datasets At Hugging Face. ttps://Huggingface.Co/Datasets/Nks9/Nks_Eye_Disease_Classification

[34] Chen PN, Lee CC, Liang CM, Pao SI, Huang KH, Lin KF. General deep learning model for detecting diabetic retinopathy. BMC bioinformatics. 2021 Nov;22:1-5.

[35] Cho H, Hwang YH, Chung JK, Lee KB, Park JS, Kim HG, Jeong JH. Deep learning ensemble method for classifying glaucoma stages using fundus photographs and convolutional neural networks. Current eye research. 2021 Oct 3;46(10):1516-24.

[36] Mehboob A, Akram MU, Alghamdi NS, Abdul Salam A. A deep learning based approach for grading of diabetic retinopathy using large fundus image dataset. Diagnostics. 2022 Dec 7;12(12):3084.

[37] Muchuchuti S, Viriri S. Retinal disease detection using deep learning techniques: a comprehensive review. Journal of Imaging. 2023 Apr 18;9(4):84.