{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,20]],"date-time":"2026-06-20T16:39:28Z","timestamp":1781973568371,"version":"3.54.5"},"reference-count":29,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2025,2,6]],"date-time":"2025-02-06T00:00:00Z","timestamp":1738800000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health (NIH)","doi-asserted-by":"publisher","award":["R01EY033973"],"award-info":[{"award-number":["R01EY033973"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Bioengineering"],"abstract":"<jats:p>Background: Diabetic retinopathy (DR) is the leading cause of blindness among working-age adults. Early detection is crucial to reducing DR-related vision loss risk but is fraught with challenges. Manual detection is labor-intensive and often misses tiny DR lesions, necessitating automated detection. Objective: We aimed to develop and validate an annotation-free deep learning strategy for the automatic detection of exudates and bleeding spots on color fundus photography (CFP) images and ultrawide field (UWF) retinal images. Materials and Methods: Three cohorts were created: two CFP cohorts (Kaggle-CFP and E-Ophtha) and one UWF cohort. Kaggle-CFP was used for algorithm development, while E-Ophtha, with manually annotated DR-related lesions, served as the independent test set. For additional independent testing, 50 DR-positive cases from both the Kaggle-CFP and UWF cohorts were manually outlined for bleeding and exudate spots. The remaining cases were used for algorithm training. A multiscale contrast-based shape descriptor transformed DR-verified retinal images into contrast fields. High-contrast regions were identified, and local image patches from abnormal and normal areas were extracted to train a U-Net model. Model performance was evaluated using sensitivity and false positive rates based on manual annotations in the independent test sets. Results: Our trained model on the independent CFP cohort achieved high sensitivities for detecting and segmenting DR lesions: microaneurysms (91.5%, 9.04 false positives per image), hemorrhages (92.6%, 2.26 false positives per image), hard exudates (92.3%, 7.72 false positives per image), and soft exudates (90.7%, 0.18 false positives per image). For UWF images, the model\u2019s performance varied by lesion size. Bleeding detection sensitivity increased with lesion size, from 41.9% (6.48 false positives per image) for the smallest spots to 93.4% (5.80 false positives per image) for the largest. Exudate detection showed high sensitivity across all sizes, ranging from 86.9% (24.94 false positives per image) to 96.2% (6.40 false positives per image), though false positive rates were higher for smaller lesions. Conclusions: Our experiments demonstrate the feasibility of training a deep learning neural network for detecting and segmenting DR-related lesions without relying on their manual annotations.<\/jats:p>","DOI":"10.3390\/bioengineering12020157","type":"journal-article","created":{"date-parts":[[2025,2,6]],"date-time":"2025-02-06T06:30:29Z","timestamp":1738823429000},"page":"157","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Geometric Self-Supervised Learning: A Novel AI Approach Towards Quantitative and Explainable Diabetic Retinopathy Detection"],"prefix":"10.3390","volume":"12","author":[{"given":"Lucas","family":"Pu","sequence":"first","affiliation":[{"name":"Department of Radiology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15260, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Oliver","family":"Beale","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Meng","sequence":"additional","affiliation":[{"name":"Department of Radiology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15260, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,2,6]]},"reference":[{"key":"ref_1","unstructured":"(2024, February 24). Diabetes Around the World in 2021. Available online: https:\/\/idf.org\/about-diabetes\/diabetes-facts-figures\/."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"226","DOI":"10.2337\/diacare.26.1.226","article-title":"Diabetic Retinopathy","volume":"26","author":"Fong","year":"2003","journal-title":"Diabetes Care"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"P66","DOI":"10.1016\/j.ophtha.2019.09.025","article-title":"Diabetic Retinopathy Preferred Practice Pattern(R)","volume":"127","author":"Flaxel","year":"2020","journal-title":"Ophthalmology"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1016\/j.ophtha.2009.06.051","article-title":"The 25-year incidence of visual impairment in type 1 diabetes mellitus the wisconsin epidemiologic study of diabetic retinopathy","volume":"117","author":"Klein","year":"2010","journal-title":"Ophthalmology"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"588","DOI":"10.1136\/bjophthalmol-2019-314336","article-title":"Estimated number of ophthalmologists worldwide (International Council of Ophthalmology update): Will we meet the needs?","volume":"104","author":"Resnikoff","year":"2020","journal-title":"Br. J. Ophthalmol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"2987","DOI":"10.4103\/ijo.IJO_1411_21","article-title":"Barriers in establishing systematic diabetic retinopathy screening through telemedicine in low- and middle-income countries","volume":"69","author":"Yim","year":"2021","journal-title":"Indian. J. Ophthalmol."},{"key":"ref_7","unstructured":"(2022, March 09). Why Is AI Adoption in Health Care Lagging?. Available online: https:\/\/www.brookings.edu\/articles\/why-is-ai-adoption-in-health-care-lagging\/."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Yang, Z., Tan, T.E., Shao, Y., Wong, T.Y., and Li, X. (2022). Classification of diabetic retinopathy: Past, present and future. Front. Endocrinol., 13.","DOI":"10.3389\/fendo.2022.1079217"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.cmpb.2018.02.016","article-title":"Microaneurysm detection using fully convolutional neural networks","volume":"158","author":"Chudzik","year":"2018","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1016\/j.ins.2017.08.050","article-title":"Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network","volume":"420","author":"Tan","year":"2017","journal-title":"Inf. Sci."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1016\/j.cmpb.2017.10.017","article-title":"An ensemble deep learning based approach for red lesion detection in fundus images","volume":"153","author":"Orlando","year":"2018","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Benzamin, A., and Chakraborty, C. (2018, January 6\u20137). Detection of Hard Exudates in Retinal Fundus Images Using Deep Learning. Proceedings of the 2018 IEEE International Conference on System, Computation, Automation and Networking (ICSCA), Pondicherry, India.","DOI":"10.1109\/ICSCAN.2018.8541246"},{"key":"ref_13","unstructured":"(2022, July 01). Diabetic Retinopathy Detection. Available online: https:\/\/www.kaggle.com\/competitions\/diabetic-retinopathy-detection\/data."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"196","DOI":"10.1016\/j.irbm.2013.01.010","article-title":"TeleOphta: Machine learning and image processing methods for teleophthalmology","volume":"34","author":"Cazuguel","year":"2013","journal-title":"IRBM"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"100512","DOI":"10.1016\/j.patter.2022.100512","article-title":"DeepDRiD: Diabetic Retinopathy-Grading and Image Quality Estimation Challenge","volume":"3","author":"Liu","year":"2022","journal-title":"Patterns"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"3300","DOI":"10.1109\/TIP.2018.2815345","article-title":"Retinal Microaneurysms Detection Using Local Convergence Index Features","volume":"27","author":"Dashtbozorg","year":"2018","journal-title":"IEEE Trans. Image Process"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Melo, T., Mendonca, A.M., and Campilho, A. (2020). Microaneurysm detection in color eye fundus images for diabetic retinopathy screening. Comput. Biol. Med., 126.","DOI":"10.1016\/j.compbiomed.2020.103995"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Sun, S., Cao, Z., Liao, D., and Lv, R. (2021). A Magnified Adaptive Feature Pyramid Network for automatic microaneurysms detection. Comput. Biol. Med., 139.","DOI":"10.1016\/j.compbiomed.2021.105000"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Eftekhari, N., Pourreza, H.R., Masoudi, M., Ghiasi-Shirazi, K., and Saeedi, E. (2019). Microaneurysm detection in fundus images using a two-step convolutional neural network. Biomed. Eng. Online, 18.","DOI":"10.1186\/s12938-019-0675-9"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1016\/j.compmedimag.2016.09.001","article-title":"A location-to-segmentation strategy for automatic exudate segmentation in colour retinal fundus images","volume":"55","author":"Liu","year":"2017","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1026","DOI":"10.1016\/j.media.2014.05.004","article-title":"Exudate detection in color retinal images for mass screening of diabetic retinopathy","volume":"18","author":"Zhang","year":"2014","journal-title":"Med. Image Anal."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1016\/j.bspc.2017.02.012","article-title":"Multiscale segmentation of exudates in retinal images using contextual cues and ensemble classification","volume":"35","author":"Fraz","year":"2017","journal-title":"Biomed. Signal Process. Control."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"024002","DOI":"10.1117\/1.JMI.4.2.024002","article-title":"Tsallis entropy and sparse reconstructive dictionary learning for exudate detection in diabetic retinopathy","volume":"4","author":"Das","year":"2017","journal-title":"J. Med. Imaging"},{"key":"ref_25","unstructured":"Piotr, C., Somshubra, M., Francesco, C., Bashir, A.-D., and Andrew, H. (2018). Exudate Segmentation Using Fully Convolutional Neural Networks and Inception Modules, ProcSPIE."},{"key":"ref_26","unstructured":"Piotr, C., Somshubra, M., Francesco, C., Bashir, A.-D., and Andrew, H. (2018). Microaneurysm Detection Using Deep Learning and Interleaved Freezing, ProcSPIE."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1016\/j.compmedimag.2016.08.001","article-title":"Automatic detection of microaneurysms in retinal fundus images","volume":"55","author":"Wu","year":"2017","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Canadian Ophthalmological Society Diabetic Retinopathy Clinical Practice Guideline Expert Committee, Hooper, P., Boucher, M.C., Cruess, A., Dawson, K.G., Delpero, W., Greve, M., Kozousek, V., Lam, W.C., and Maberley, D.A. (2012). Canadian Ophthalmological Society Evidence-based Clinical Practice Guidelines for the Management of Diabetic Retinopathy\u2014Executive summary. Can. J. Ophthalmol., 47, 91\u201396.","DOI":"10.1016\/j.jcjo.2012.01.022"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"965","DOI":"10.1111\/j.1464-5491.2003.01077.x","article-title":"Grading and disease management in national screening for diabetic retinopathy in England and Wales","volume":"20","author":"Harding","year":"2003","journal-title":"Diabet. Med."}],"container-title":["Bioengineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2306-5354\/12\/2\/157\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T16:28:02Z","timestamp":1760027282000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2306-5354\/12\/2\/157"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,6]]},"references-count":29,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,2]]}},"alternative-id":["bioengineering12020157"],"URL":"https:\/\/doi.org\/10.3390\/bioengineering12020157","relation":{},"ISSN":["2306-5354"],"issn-type":[{"value":"2306-5354","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,6]]}}}