Samia Akhtar; Shabib Aftab; Munir Ahmad; Asma Akhtar

Diabetic Retinopathy Severity Grading Using Transfer Learning Techniques

PDF (724KB), PP.41-53

Views: 0 Downloads: 0

Author(s)

Samia Akhtar ¹ Shabib Aftab ^1,* Munir Ahmad ² Asma Akhtar ¹

1. Department of Computer Science, Virtual University of Pakistan, Lahore, Pakistan

2. School of Computer Science, NCBA&E, Lahore, Pakistan

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2024.06.04

Received: 17 Feb. 2024 / Revised: 3 Jul. 2024 / Accepted: 14 Aug. 2024 / Published: 8 Dec. 2024

Index Terms

Diabetic Retinopathy, Transfer Learning, Artificial Intelligence, Deep Learning, Diabetes

Abstract

Diabetic Retinopathy is a severe eye condition originating as a result of long term diabetes mellitus. Timely detection is essential to prevent it from progressing to more advanced stages. Manual detection of DR is labor-intensive and time-consuming, requiring expertise and extensive image analysis. Our research aims to develop a robust and automated deep learning model to assist healthcare professionals by streamlining the detection process and improving diagnostic accuracy. This research proposes a multi-classification framework using Transfer Learning for diabetic retinopathy grading among diabetic patients. An image based dataset, APTOS 2019 Blindness Detection, is utilized for our model training and testing. Our methodology involves three key preprocessing steps: 1) Cropping to remove extraneous background regions, 2) Contrast enhancement using CLAHE (Contrast Limited Adaptive Histogram Equalization) and 3) Resizing to a consistent dimension of 224x224x3. To address class imbalance, we applied SMOTE (Synthetic Minority Over-sampling Technique) for balancing the dataset. Data augmentation techniques such as rotation, zooming, shifting, and brightness adjustment are used to further enhance the model's generalization. The dataset is split to a 70:10:20 ratios for training, validation and testing. For classification, EfficientNetB3 and Xception, two transfer learning models, are used after fine-tuning which includes addition of dense, dropout and fully connected layers. Hyper parameters such as batch size, no. of epochs, optimizer etc were adjusted prioir model training. The performance of our model is evaluated using various performance metrics including accuracy, specificity, sensitivity and others. Results reveal the highest test accuracy of 95.16% on the APTOS dataset for grading diabetic retinopathy into five classes using the EfficientNetB3 model followed by a test accuracy of 92.66% using Xception model. Our top-performing model, EfficientNetB3, was compared against various state-of-the-art approaches, including DenseNet-169, hybrid models, and ResNet-50, where our model outperformed all these methodologies.

Cite This Paper

Samia Akhtar, Shabib Aftab, Munir Ahmad, Asma Akhtar, "Diabetic Retinopathy Severity Grading Using Transfer Learning Techniques", International Journal of Engineering and Manufacturing (IJEM), Vol.14, No.6, pp. 41-53, 2024. DOI:10.5815/ijem.2024.06.04

Reference

[1]Skouta, Ayoub, Abdelali Elmoufidi, Said Jai-Andaloussi, and Ouail Ochetto. "Automated binary classification of diabetic retinopathy by convolutional neural networks." In Advances on Smart and Soft Computing: Proceedings of ICACI, Springer Singapore, pp. 177-187, 2021, doi: https://doi.org/10.1007/978-981-15-6048-4_16.
[2]Ahmed, Usama, Ghassan F. Issa, Muhammad Adnan Khan, Shabib Aftab, Muhammad Farhan Khan, Raed AT Said, Taher M. Ghazal, and Munir Ahmad. "Prediction of diabetes empowered with fused machine learning." IEEE Access 10, pp: 8529-8538, 2022, doi: https://doi.org/10.1109/ACCESS.2022.3142097
[3]Kassani, Sara Hosseinzadeh, Peyman Hosseinzadeh Kassani, Reza Khazaeinezhad, Michal J. Wesolowski, Kevin A. Schneider, and Ralph Deters. "Diabetic retinopathy classification using a modified xception architecture." In 2019 IEEE international symposium on signal processing and information technology (ISSPIT), IEEE, pp. 1-6, 2019, doi: https://doi.org/10.1109/ISSPIT47144.2019.9001846.
[4]Le, David, Minhaj Alam, Cham K. Yao, Jennifer I. Lim, Yi-Ting Hsieh, Robison VP Chan, Devrim Toslak, and Xincheng Yao. "Transfer learning for automated OCTA detection of diabetic retinopathy." Translational Vision Science & Technology 9, no. 2, pp. 30-35, 2020, doi: https://doi.org/10.1167%2Ftvst.9.2.35.
[5]Aswathi, T., T. R. Swapna, and S. Padmavathi. "Transfer learning approach for grading of diabetic retinopathy." Journal of Physics: Conference Series, vol. 1767, no. 1, pp. 012033, 2021, doi: 10.1088/1742-6596/1767/1/012033.
[6]Butt, Muhammad Mohsin, DNF Awang Iskandar, Sherif E. Abdelhamid, Ghazanfar Latif, and Runna Alghazo. "Diabetic retinopathy detection from fundus images of the eye using hybrid deep learning features." Diagnostics 12, no. 7, pp. 1607, 2022, doi: https://doi.org/10.3390%2Fdiagnostics12071607.
[7]Vaibhavi, P. M., and R. Manjesh. "Binary classification of diabetic retinopathy detection and web application." International Journal of Research in Engineering, Science and Management 4, no. 7, pp. 142-145, 2021, doi: https://journal.ijresm.com/index.php/ijresm/article/view/1000.
[8]El Houby, Enas MF. "Using transfer learning for diabetic retinopathy stage classification." Applied Computing and Informatics, 2021, doi: https://doi.org/10.1108/ACI-07-2021-0191.
[9]Qiao, Lifeng, Ying Zhu, and Hui Zhou. "Diabetic retinopathy detection using prognosis of microaneurysm and early diagnosis system for non-proliferative diabetic retinopathy based on deep learning algorithms." IEEE Access 8, pp. 104292-104302, 2020, doi: https://doi.org/10.1109/ACCESS.2020.2993937.
[10]Sundar, Sumod, and S. Sumathy. "Classification of Diabetic Retinopathy disease levels by extracting topological features using Graph Neural Networks." IEEE Access 11, pp. 51435-51444, 2023, doi: https://doi.org/10.1109/ACCESS.2023.3279393.
[11]Gao, Zhiyuan, Xiangji Pan, Ji Shao, Xiaoyu Jiang, Zhaoan Su, Kai Jin, and Juan Ye. "Automatic interpretation and clinical evaluation for fundus fluorescein angiography images of diabetic retinopathy patients by deep learning." British Journal of Ophthalmology 107, no. 12, pp. 1852-1858, 2023. doi: http://dx.doi.org/10.1136/bjo-2022-321472.
[12]Salvi, Raj Sunil, Shreyas Rajesh Labhsetwar, Piyush Arvind Kolte, Veerasai Subramaniam Venkatesh, and Alistair Michael Baretto. "Predictive analysis of diabetic retinopathy with transfer learning." 4th Biennial International Conference on Nascent Technologies in Engineering (ICNTE), IEEE, pp. 1-6, 2021. doi: https://doi.org/10.1109/ICNTE51185.2021.9487789.
[13]Elsharkawy, Mohamed, Ahmed Sharafeldeen, Ahmed Soliman, Fahmi Khalifa, Mohammed Ghazal, Eman El-Daydamony, Ahmed Atwan, Harpal Singh Sandhu, and Ayman El-Baz. "A novel computer-aided diagnostic system for early detection of diabetic retinopathy using 3D-OCT higher-order spatial appearance model." Diagnostics 12, no. 2, p. 461, 2022. doi: https://doi.org/10.3390/diagnostics12020461.
[14]Li, Feng, Yuguang Wang, Tianyi Xu, Lin Dong, Lei Yan, Minshan Jiang, Xuedian Zhang, Hong Jiang, Zhizheng Wu, and Haidong Zou. "Deep learning-based automated detection for diabetic retinopathy and diabetic macular oedema in retinal fundus photographs." Eye 36, no. 7, pp: 1433-1441, 2022, doi: https://doi.org/10.1038/s41433-021-01552-8.
[15]Wang, Xiaoling, Zexuan Ji, Xiao Ma, Ziyue Zhang, Zuohuizi Yi, Hongmei Zheng, Wen Fan, and Changzheng Chen. "Automated Grading of Diabetic Retinopathy with Ultra‐Widefield Fluorescein Angiography and Deep Learning." Journal of Diabetes Research, no. 1, pp. 2611250, 2021, doi: https://doi.org/10.1155/2021/2611250.
[16]Gangwar, Akhilesh Kumar, and Vadlamani Ravi. "Diabetic retinopathy detection using transfer learning and deep learning." In Evolution in Computational Intelligence: Frontiers in Intelligent Computing: Theory and Applications (FICTA 2020), Springer Singapore, Volume 1, pp. 679-689, 2021, doi: https://doi.org/10.1007/978-981-15-5788-0_64
[17]Chen, Ping-Nan, Chia-Chiang Lee, Chang-Min Liang, Shu-I. Pao, Ke-Hao Huang, and Ke-Feng Lin. "General deep learning model for detecting diabetic retinopathy." BMC bioinformatics 22, pp: 1-15, 2021, doi: https://doi.org/10.1186/s12859-021-04005-x
[18]Sikder, Niloy, Mehedi Masud, Anupam Kumar Bairagi, Abu Shamim Mohammad Arif, Abdullah-Al Nahid, and Hesham A. Alhumyani. "Severity classification of diabetic retinopathy using an ensemble learning algorithm through analyzing retinal images." Symmetry 13, no. 4, p. 670, 2021, doi: https://doi.org/10.3390/sym13040670.
[19]Hagos, Misgina Tsighe, and Shri Kant. "Transfer learning based detection of diabetic retinopathy from small dataset." arXiv, pp.1905.07203, 2019, doi: https://doi.org/10.48550/arXiv.1905.07203.
[20]Shi, Danli, Weiyi Zhang, Shuang He, Yanxian Chen, Fan Song, Shunming Liu, Ruobing Wang, Yingfeng Zheng, and Mingguang He. "Translation of color fundus photography into fluorescein angiography using deep learning for enhanced diabetic retinopathy screening." Ophthalmology science 3, no. 4, 100401, 2023, doi: https://doi.org/10.1016/j.xops.2023.100401.
[21]Menaouer, Brahami, Zoulikha Dermane, Nour El Houda Kebir, and Nada Matta. "Diabetic retinopathy classification using hybrid deep learning approach." SN Computer Science 3, no. 5, p. 357, 2022, doi: https://doi.org/10.1007/s42979-022-01240-8.
[22]Alyoubi, Wejdan L., Maysoon F. Abulkhair, and Wafaa M. Shalash. "Diabetic retinopathy fundus image classification and lesions localization system using deep learning." Sensors 21, no. 11, p. 3704, 2021, doi: https://doi.org/10.3390/s21113704.
[23]Abbood, Saif Hameed, Haza Nuzly Abdull Hamed, Mohd Shafry Mohd Rahim, Amjad Rehman, Tanzila Saba, and Saeed Ali Bahaj. "Hybrid retinal image enhancement algorithm for diabetic retinopathy diagnostic using deep learning model." IEEE Access 10, pp. 73079-73086, 2022, doi: https://doi.org/10.1109/ACCESS.2022.3189374.
[24]Shaban, Mohamed, Zeliha Ogur, Ali Mahmoud, Andrew Switala, Ahmed Shalaby, Hadil Abu Khalifeh, Mohammed Ghazal et al. "A convolutional neural network for the screening and staging of diabetic retinopathy." Plos one 15, no. 6, p. e0233514, 2020, doi: http://dx.doi.org/10.1371/journal.pone.0233514.
[25]Lin, Chun-Ling, and Kun-Chi Wu. "Development of revised ResNet-50 for diabetic retinopathy detection." BMC bioinformatics 24, no. 1, pp. 157, 2023, doi: https://doi.org/10.1186/s12859-023-05293-1.
[26]Mushtaq, Gazala, and Farheen Siddiqui. "Detection of diabetic retinopathy using deep learning methodology." In IOP conference series: materials science and engineering, IOP Publishing, vol. 1070, no. 1, p. 012049, 2021, doi: 10.1088/1757-899X/1070/1/012049.
[27]Raja Kumar, R., R. Pandian, T. Prem Jacob, A. Pravin, and P. Indumathi. "Detection of diabetic retinopathy using deep convolutional neural networks." In Computational Vision and Bio-Inspired Computing: ICCVBIC, Springer Singapore, pp. 415-430, 2021, doi: https://doi.org/10.1007/978-981-33-6862-0_34.
[28]Liu, Hao, Keqiang Yue, Siyi Cheng, Chengming Pan, Jie Sun, and Wenjun Li. "Hybrid model structure for diabetic retinopathy classification." Journal of Healthcare Engineering, no. 1, pp. 8840174, 2020, doi: https://doi.org/10.1155/2020/8840174.
[29]Bhardwaj, Charu, Shruti Jain, and Meenakshi Sood. "Transfer learning based robust automatic detection system for diabetic retinopathy grading." Neural Computing and Applications 33, no. 20, pp. 13999-14019, 2021, doi: https://doi.org/10.1007/s00521-021-06042-2.
[30]https://www.kaggle.com/competitions/aptos2019-blindness-detection.
[31]Wang, Juan, Yujing Bai, and Bin Xia. "Feasibility of diagnosing both severity and features of diabetic retinopathy in fundus photography." IEEE access 7, pp: 102589-102597, 2019, doi: https://doi.org/10.1109/ACCESS.2019.2930941.
[32]Masood, Sarfaraz, Tarun Luthra, Himanshu Sundriyal, and Mumtaz Ahmed. "Identification of diabetic retinopathy in eye images using transfer learning." International conference on computing, communication and automation (ICCCA), IEEE, pp. 1183-1187, 2017, doi: https://doi.org/10.1016/j.procs.2020.03.400.
[33]Li, Xiaogang, Tiantian Pang, Biao Xiong, Weixiang Liu, Ping Liang, and Tianfu Wang. "Convolutional neural networks based transfer learning for diabetic retinopathy fundus image classification." 10th international congress on image and signal processing, biomedical engineering and informatics (CISP-BMEI), IEEE, pp. 1-11, 2017, doi: https://doi.org/10.1109/CISP-BMEI.2017.8301998.
[34]Thota, Narayana Bhagirath, and Doshna Umma Reddy. "Improving the accuracy of diabetic retinopathy severity classification with transfer learning." 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), IEEE, pp. 1003-1006, 2020, doi: https://doi.org/10.1109/MWSCAS48704.2020.9184473.
[35]Kotiyal, Bina, and Heman Pathak. "Diabetic retinopathy binary image classification using PySpark." International Journal of Mathematical, Engineering and Management Sciences 7, no. 5, p. 624, 2022, doi: 10.33889/IJMEMS.2022.7.5.041.
[36]Patel, Rupa, and Anita Chaware. "Transfer learning with fine-tuned MobileNetV2 for diabetic retinopathy." In 2020 international conference for emerging technology (INCET), IEEE, pp. 1-4, 2020, doi: https://doi.org/10.1109/INCET49848.2020.9154014.
[37]Bora, Ashish, Siva Balasubramanian, Boris Babenko, Sunny Virmani, Subhashini Venugopalan, Akinori Mitani, Guilherme de Oliveira Marinho et al. "Predicting the risk of developing diabetic retinopathy using deep learning." The Lancet Digital Health 3, no. 1, pp: e10-e19, 2021, doi: https://doi.org/10.1016/S2589-7500(20)30250-8.
[38]Martínez-Murcia, Francisco Jesús, Andrés Ortiz-García, Javier Ramírez, Juan Manuel Górriz-Sáez, and Ricardo Cruz. "Deep Residual Transfer Learning for Automatic Diabetic Retinopathy Grading.", 2021, doi: https://dx.doi.org/10.1016/j.neucom.2020.04.148.
[39]Sebti, Riad, Siham Zroug, Laid Kahloul, and Saber Benharzallah. "A deep learning approach for the diabetic retinopathy detection." In The Proceedings of the International Conference on Smart City Applications, Cham: Springer International Publishing, pp. 459-469, 2021, doi: https://doi.org/10.1007/978-3-030-94191-8_37.
[40]Çinarer, Gökalp, Kazım Kiliç, and Tuba Parlar. "A Deep Transfer Learning Framework for the Staging of Diabetic Retinopathy." Journal of Scientific Reports-A 051, pp. 106-119, 2022.
[41]AbdelMaksoud, Eman, Sherif Barakat, and Mohammed Elmogy. "A computer-aided diagnosis system for detecting various diabetic retinopathy grades based on a hybrid deep learning technique." Medical & Biological Engineering & Computing 60, no. 7, pp: 2015-2038, 2022, doi: https://doi.org/10.1007/s11517-022-02564-6.
[42]Al-Smadi, Mohammed, Mahmoud Hammad, Qanita Bani Baker, and A. Sa’ad. "A transfer learning with deep neural network approach for diabetic retinopathy classification." International Journal of Electrical and Computer Engineering 11, no. 4, p. 3492, 2021, doi: 10.11591/ijece.v11i4.pp3492-3501.
[43]Ebrahimi, Behrouz, David Le, Mansour Abtahi, Albert K. Dadzie, Jennifer I. Lim, RV Paul Chan, and Xincheng Yao. "Optimizing the OCTA layer fusion option for deep learning classification of diabetic retinopathy." Biomedical Optics Express 14, no. 9, pp. 4713-4724, 2023, doi: https://doi.org/10.1364/BOE.495999.
[44]Shakibania, Hossein, Sina Raoufi, Behnam Pourafkham, Hassan Khotanlou, and Muharram Mansoorizadeh. "Dual branch deep learning network for detection and stage grading of diabetic retinopathy." Biomedical Signal Processing and Control 93, p. 106168, 2024, doi: https://doi.org/10.1016/j.bspc.2024.106168.
[45]Jabbar, Ayesha, Hannan Bin Liaqat, Aftab Akram, Muhammad Usman Sana, Irma Domínguez Azpíroz, Isabel De La Torre Diez, and Imran Ashraf. "A Lesion-Based Diabetic Retinopathy Detection Through Hybrid Deep Learning Model." IEEE Access, 2024, doi: https://doi.org/10.1109/ACCESS.2024.3373467.
[46]Singh, Devendra, and Dinesh C. Dobhal. "A Deep Learning-based Transfer Learning Approach Fine-Tuned for Detecting Diabetic Retinopathy." Procedia Computer Science 233, pp: 444-453, 2024, doi: https://doi.org/10.1016/j.procs.2024.03.234.
[47]Romero-Oraá, Roberto, María Herrero-Tudela, María I. López, Roberto Hornero, and María García. "Attention-based deep learning framework for automatic fundus image processing to aid in diabetic retinopathy grading." Computer Methods and Programs in Biomedicine 249, p. 108160, 2024, doi: https://doi.org/10.1016/j.cmpb.2024.108160.
[48]Mutawa, A. M., Khalid Al-Sabti, Seemant Raizada, and Sai Sruthi. "A Deep Learning Model for Detecting Diabetic Retinopathy Stages with Discrete Wavelet Transform." Applied Sciences 14, no. 11, p. 4428, 2024, doi: https://doi.org/10.3390/app14114428.
[49]Chilukoti, Sai Venkatesh, Anthony S. Maida, and Xiali Hei. "Diabetic retinopathy detection using transfer learning from pre-trained convolutional neural network models." IEEE J Biomed Heal Informatics 20, pp: 1-10, 2022, doi: https://dx.doi.org/10.36227/techrxiv.18515357.v1.
[50]Boruah, Swagata, Archit Dehloo, Prajul Gupta, Manas Ranjan Prusty, and A. Balasundaram. "Gaussian blur masked resnet2. 0 architecture for diabetic retinopathy detection." Computers, Materials and Continua 75, no. 1, pp: 927-942, 2023, doi: https://doi.org/10.32604/cmc.2023.035143.

International Journal of Engineering and Manufacturing (IJEM)