Abstract
Purpose
This study aims to develop a deep-learning-based software capable of detecting and differentiating microaneurysms (MAs) as hyporeflective or hyperreflective on structural optical coherence tomography (OCT) images in patients with non-proliferative diabetic retinopathy (NPDR).
Methods
A retrospective cohort of 249 patients (498 eyes) diagnosed with NPDR was analysed. Structural OCT scans were obtained using the Heidelberg Spectralis HRA + OCT device. Manual segmentation of MAs was performed by five masked readers, with an expert grader ensuring consistent labeling. Two deep learning models, YOLO (You Only Look Once) and DETR (DEtection TRansformer), were trained using the annotated OCT images. Detection and classification performance were evaluated using the area under the receiver operating characteristic (ROC) curves.
Results
The YOLO model performed poorly with an AUC of 0.35 for overall MA detection, with AUCs of 0.33 and 0.24 for hyperreflective and hyporeflective MAs, respectively. The DETR model had an AUC of 0.86 for overall MA detection, but AUCs of 0.71 and 0.84 for hyperreflective and hyporeflective MAs, respectively. Post-hoc review revealed that discrepancies between automated and manual grading were often due to the automated method’s selection of normal retinal vessels.
Conclusions
The choice of deep learning model is critical to achieving accuracy in detecting and classifying MAs in structural OCT images. An automated approach may assist clinicians in the early detection and monitoring of diabetic retinopathy, potentially improving patient outcomes.
Similar content being viewed by others
Log in or create a free account to read this content
Gain free access to this article, as well as selected content from this journal and more on nature.com
or
Data availability
Data that support the findings of this study are available upon reasonable request to the corresponding author.
References
Wong TY, Sabanayagam C. Strategies to tackle the global burden of diabetic retinopathy: from epidemiology to artificial intelligence. Ophthalmologica. 2020;243:9–20.
Ansari P, Tabasumma N, Snigdha NN, Siam NH, Panduru RVNRS, et al. Diabetic retinopathy: an overview on mechanisms, pathophysiology and pharmacotherapy. Diabetology. 2022;3:159–75.
Forrester JV, Kuffova L, Delibegovic M. The role of inflammation in diabetic retinopathy. Front Immunol. 2020;11:1–22.
Borrelli E, Sacconi R, Brambati M, Bandello F, Querques G. In vivo rotational three-dimensional OCTA analysis of microaneurysms in the human diabetic retina. Sci Rep. 2019;9:16789. https://doi.org/10.1038/s41598-019-53357-1.
Akram MU, Khalid S, Khan SA. Identification and classification of microaneurysms for early detection of diabetic retinopathy. Pattern Recognit. 2013;46:107–16.
Horii T, Murakami T, Nishijima K, Sakamoto A, Ota M, Yoshimura N. Optical coherence tomographic characteristics of microaneurysms in diabetic retinopathy. Am J Ophthalmol. 2010;150:840–8.
Querques G, Borrelli E, Battista M, Sacconi R, Bandello F Optical coherence tomography angiography in diabetes: focus on microaneurysms. Eye. 2020;35 https://pubmed.ncbi.nlm.nih.gov/32887935/.
Borrelli E, Battista M, Sacconi R, Querques G, Bandello F. Optical coherence tomography angiography in diabetes. Asia Pac J Ophthalmol. 2021;10.
Borrelli E, Sacconi R, Parravano M, Costanzo E, Querques L, Battista M, et al. OCTA assessment of the diabetic macula: a comparison study among different algorithms. Retina. 2021.
Kaizu Y, Nakao S, Wada I, Arima M, Yamaguchi M, Ishikawa K, et al. Microaneurysm imaging using multiple En face OCT angiography image averaging: morphology and visualization. Ophthalmol Retina. 2020.
Karst SG, Salas M, Hafner J, Scholda C, Vogl W-D, Drexler W, et al. Three-dimensional analysis of retinal microaneurysms with adaptive optics optical coherence tomography. Retina. 2019.
Parravano M, De Geronimo D, Scarinci F, Querques L, Virgili G, Simonett JM, et al. Diabetic microaneurysms internal reflectivity on spectral-domain optical coherence tomography and optical coherence tomography angiography detection. Am J Ophthalmol.2017;179:90–6. https://linkinghub.elsevier.com/retrieve/pii/S0002939417301903.
Parravano M, De Geronimo D, Scarinci F, Virgili G, Querques L, Varano M, et al. Progression of diabetic microaneurysms according to the internal reflectivity on structural optical coherence tomography and visibility on optical coherence tomography angiography. Am J Ophthalmol. 2019;198:8–16. http://www.ncbi.nlm.nih.gov/pubmed/30308201.
Arrigo A, Teussink M, Aragona E, Bandello F, Battaglia Parodi M. MultiColor imaging to detect different subtypes of retinal microaneurysms in diabetic retinopathy. Eye. 2021;35:277–81.
Parravano M, De Geronimo D, Scarinci F, Querques L, Virgili G, Simonett JM, et al. Diabetic microaneurysms internal reflectivity on spectral-domain optical coherence tomography and optical coherence tomography angiography detection. Am J Ophthalmol. 2017;179:90–6.
Sun Z, Yang D, Tang Z, Ng DS, Cheung CY. Optical coherence tomography angiography in diabetic retinopathy: an updated review. Eye. 2021;35:149–61.
Borrelli E, Grosso D, Barresi C, Lari G, Sacconi R, Senni C, et al. Long-term visual outcomes and morphologic biomarkers of vision loss in eyes with diabetic macular edema treated with anti-VEGF Therapy. Am J Ophthalmol. 2021. https://pubmed.ncbi.nlm.nih.gov/34509431/.
Vujosevic S, Aldington SJ, Silva P, Hernández C, Scanlon P, Peto T, et al. Screening for diabetic retinopathy: new perspectives and challenges. Lancet Diabetes Endocrinol. 2020;8:337–47.
Oakley JD, Verdooner S, Russakoff DB, Brucker AJ, Seaman J, Sahni J, et al. Quantitative assessment of automated optical coherence tomography image analysis using a home-based device for self-monitoring neovascular age-related macular degeneration. Retina. 2022.
Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation BT - medical image computing and computer-assisted intervention – MICCAI 2015. In: Navab N, Hornegger J, Wells WM, Frangi AF, editors. Cham: Springer International Publishing; 2015. p. 234–41.
Oakley JD, Sodhi SK, Russakoff DB, Choudhry N. Automated Deep Learning-based Multi-class Fluid Segmentation in Swept-Source Optical Coherence Tomography Images. 2020. https://doi.org/10.1101/2020.09.01.278259.
Borrelli E, Oakley JD, Iaccarino G, Russakoff DB, Battista M, Grosso D, et al. Deep-learning based automated quantification of critical optical coherence tomography features in neovascular age-related macular degeneration. Eye. 2023.
Ricardi F, Oakley J, Russakoff D, Boscia G, Caselgrandi P, Gelormini F, et al. Validation of a deep learning model for automatic detection and quantification of five OCT critical retinal features associated with neovascular age-related macular degeneration. Br J Ophthalmol. 2024: bjo-2023-324647.
Redmon J, Divvala S, Girshick R, Farhadi A. You only look once: unified, real-time object detection. http://pjreddie.com/yolo/.
Bochkovskiy A, Wang C-Y, Liao H-YM. YOLOv4: optimal speed and accuracy of object detection. 2020. http://arxiv.org/abs/2004.10934.
Zhu X, Su W, Lu L, Li B, Wang X, Dai J. Deformable DETR: deformable transformers for end-to-end object detection. 2020. http://arxiv.org/abs/2010.04159.
YOUDEN WJ. Index for rating diagnostic tests. Cancer. 1950;3:32–5.
Almasi R, Vafaei A, Kazeminasab E, Rabbani H. Automatic detection of microaneurysms in optical coherence tomography images of retina using convolutional neural networks and transfer learning. Sci Rep. 2022;12:13975.
Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016;316:2402–10. https://jamanetwork.com/journals/jama/fullarticle/2588763.
Abràmoff MD, Lou Y, Erginay A, Clarida W, Amelon R, Folk JC, et al. Improved automated detection of diabetic retinopathy on a publicly available dataset through integration of deep learning. Invest Ophthalmol Vis Sci. 2016;57:5200–6.
Pratt H, Coenen F, Broadbent DM, Harding SP, Zheng Y. Convolutional neural networks for diabetic retinopathy. In: Procedia computer science. Elsevier B.V. 2016;90:200–5.
Abràmoff MD, Lavin PT, Birch M, Shah N, Folk JC. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit Med. 2018;1.
Parravano M, De Geronimo D, Scarinci F, Virgili G, Querques L, Varano M, et al. Progression of diabetic microaneurysms according to the internal reflectivity on structural optical coherence tomography and visibility on optical coherence tomography angiography. Am J Ophthalmol. 2019;198:8–16.
Zhang L, Van Dijk EHC, Borrelli E, Fragiotta S, Breazzano MP. OCT and OCT angiography update: clinical application to age-related macular degeneration, central serous chorioretinopathy, macular telangiectasia, and diabetic retinopathy. Diagnostics. 2023;13:232.
Author information
Authors and Affiliations
Contributions
All authors contributed to the conceptualization, writing, and final review of the manuscript.
Corresponding author
Ethics declarations
Competing interests
Prof Borrelli is a member of the Eye editorial board. Jonathan D. Oakley: Employees (Voxeleron LLC), Patent (Voxeleron LLC); Daniel B. Russakoff: Employees (Voxeleron LLC), Patent (Voxeleron LLC); Michele Reibaldi is a consultant/advisor for: Abbvie, Apellis, Bayer, Novartis, Roche. Enrico Borrelli is a consultant/advisor for: Abbvie, Apellis, Bayer, Roche, Zeiss. The other authors have no disclosures.
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
Neri, G., Sharma, S., Ghezzo, B. et al. Deep learning model for automatic detection of different types of microaneurysms in diabetic retinopathy. Eye 39, 570–577 (2025). https://doi.org/10.1038/s41433-024-03585-1
Received:
Revised:
Accepted:
Published:
Version of record:
Issue date:
DOI: https://doi.org/10.1038/s41433-024-03585-1
