Remote Healthcare Diabetic Retinopathy Detection Using Deep Learning
Keywords:
OpenCV, Convolutional Neural Networks (CNN), Keras, Diabetic Retinopathy (DR), Deep Neural Networks, blood vesselsAbstract
High blood glucose levels are a hallmark of diabetes mellitus (DM), a metabolic disease. This can give rise to a range of complications, with Diabetic Retinopathy (DR) being among them. DR can impair vision and, if not addressed, may lead to a loss of eyesight. Symptoms include aberrant blood vessels, fluid leaks, exudates, haemorrhages, and retinal microaneurysms. With the advancement of technology, medical imaging has become one of the most useful tools for detecting and treating disorders. Nonetheless, even with these advancements, the procedure is difficult, time-consuming, and not always correct. With more modern technologies, such as Deep Neural Networks, there is a good chance that we can improve accuracy and efficiency while attempting to identify DR using an automated image analysis system. This article aims to create an automated knowledge model for identifying critical characteristics associated with DR from fundus images using OpenCV, Convolutional Neural Networks (CNNs), and Keras. The main phases in this project are to collect and preprocess a diverse collection of fundus photos, annotate them to show the severity of DR, and build a CNN model with Keras. This project requires training the model on this data, monitoring its performance, and evaluating it using measures such as accuracy, precision, and recall. Building trust among healthcare providers requires interpreting the model's predictions and illustrating its decision-making process. Furthermore, the initiative must assure adherence to healthcare legislation and ethical guidelines. Working with medical specialists to assess the model's accuracy and clinical applicability is an important step in this process. Finally, this study has the potential to transform DR diagnosis, making it more efficient and precise, and so improving patient outcomes. Continuous data collection and model development are required to ensure continued progress in this vital healthcare topic.
References
Ahsan H. Diabetic retinopathy–biomolecules and multiple pathophysiology. Diabetes Metab Syndr: Clinical Research & Reviews. 2015 Jan 1; 9(1): 51–4.
Zago GT, Andreão RV, Dorizzi B, Salles EO. Diabetic retinopathy detection using red lesion localization and convolutional neural networks. Comput Biol Med. 2020 Jan 1; 116: 103537.
Harangi B, Toth J, Baran A, Hajdu A. Automatic screening of fundus images using a combination of convolutional neural network and hand-crafted features. In 2019 41st Annual international conference of the IEEE engineering in medicine and biology society (EMBC). 2019 Jul 23; 2699–2702.
Li X, Hu X, Yu L, Zhu L, Fu CW, Heng PA. CANet: cross-disease attention network for joint diabetic retinopathy and diabetic macular edema grading. IEEE Trans Med Imaging. 2019 Nov 6; 39(5): 1483–93.
Schellini SA, Carvalho GM, Rendeiro FS, Padovani CR, Hirai FE. Prevalence of diabetes and diabetic retinopathy in a Brazilian population. Ophthalmic Epidemiol. 2014 Feb 1; 21(1): 33–8.
El-Bab MF, Shawky N, Al-Sisi A, Akhtar M. Retinopathy and risk factors in diabetic patients from Al-Madinah Al-Munawarah in the Kingdom of Saudi Arabia. Clin Ophthalmol. 2012 Feb 17: 269–76.
Atwany MZ, Sahyoun AH, Yaqub M. Deep learning techniques for diabetic retinopathy classification: A survey. IEEE Access. 2022 Mar 8; 10: 28642–55.
Chetoui M, Akhloufi MA. Explainable end-to-end deep learning for diabetic retinopathy detection across multiple datasets. J Med Imaging. 2020 Jul 1; 7(4): 044503.
Soomro TA, Afifi AJ, Gao J, Hellwich O, Khan MA, Paul M, Zheng L. Boosting sensitivity of a retinal vessel segmentation algorithm with convolutional neural network. In 2017 IEEE International Conference on Digital Image Computing: Techniques and Applications (DICTA). 2017 Nov 29; 1–8.
Shekelle PG, Morton SC, Keeler EB. Costs and benefits of health information technology. Evid Rep Technol Assess. 2006 Apr; 1(132): 1–71.
Wilkinson CP, Ferris III FL, Klein RE, Lee PP, Agardh CD, Davis M, Dills D, Kampik A, Pararajasegaram R, Verdaguer JT, Global Diabetic Retinopathy Project Group. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology. 2003 Sep 1; 110(9): 1677–82.
Porwal P, Pachade S, Kamble R, Kokare M, Deshmukh G, Sahasrabuddhe V, Meriaudeau F. Indian diabetic retinopathy image dataset (IDRiD): a database for diabetic retinopathy screening research. Data. 2018 Jul 10; 3(3): 25.
Gass JD, Agarwal A, Lavina AM, Tawansy KA. Focal inner retinal hemorrhages in patients with drusen: an early sign of occult choroidal neovascularization and chorioretinal anastomosis. Retina. 2003 Dec 1; 23(6): 741–51.
Soomro TA, Gao J, Khan T, Hani AF, Khan MA, Paul M. Computerised approaches for the detection of diabetic retinopathy using retinal fundus images: a survey. Pattern Anal Appl. 2017 Nov; 20(4): 927–61.
Grubbs FE. Errors of measurement, precision, accuracy and the statistical comparison of measuring instruments. Technometrics. 1973 Feb 1; 15(1): 53–66.
Sharma M, Sharma S, Singh G. Performance analysis of statistical and supervised learning techniques in stock data mining. Data. 2018 Nov 24; 3(4): 54.
Abràmoff MD, Reinhardt JM, Russell SR, Folk JC, Mahajan VB, Niemeijer M, Quellec G. Automated early detection of diabetic retinopathy. Ophthalmology. 2010 Jun 1; 117(6): 1147–54.
Usman I, Almejalli KA. Intelligent automated detection of microaneurysms in fundus images using feature-set tuning. IEEE Access. 2020 Apr 3; 8: 65187–96.
Soomro TA, Afifi AJ, Shah AA, Soomro S, Baloch GA, Zheng L, Yin M, Gao J. Impact of image enhancement technique on CNN model for retinal blood vessels segmentation. IEEE Access. 2019 Oct 30; 7: 158183–97.
