{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T13:55:27Z","timestamp":1786974927207,"version":"build-2736575974"},"reference-count":45,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T00:00:00Z","timestamp":1771891200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T00:00:00Z","timestamp":1774569600000},"content-version":"vor","delay-in-days":31,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Thailand\u2019s Education Hub for the Southern Region of ASEAN Countries for Ph.D. Students"},{"name":"The NSRF via the Program Management Unit for Human Resources & Institutional Development, Research and Innovation","award":["B04G640070"],"award-info":[{"award-number":["B04G640070"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Accurate fovea segmentation in fundus images is a critical step in diabetic retinopathy screening; however, it remains a challenging task due to the indistinct boundaries of the fovea. Beyond simple localization, precise segmentation offers essential clinical value for Diabetic Macular Edema (DME) management, as treatment decisions\u2013specifically the choice between intravitreal anti-VEGF injection for center-involved DME and laser therapy for extrafoveal edema\u2013depend on the accurate delineation of the foveal region. While existing methods often rely on increasing model architecture complexity, the potential of anatomical context within the training process remains under-explored. This paper presents a data-centric approach that leverages contextual information to robustly identify the fovea. We demonstrate that progressively incorporating key anatomical landmarks\u2013the optic disc, retina, and blood vessels\u2013into training labels significantly enhances fovea detection. To facilitate this, we developed IDRiD-RETA-FV, a meticulously annotated dataset comprising 81 images (54 training, 27 testing) with complete anatomical structures (inter-observer F1=0.98), and introduce MNv4Fovea, a framework designed to explicitly exploit these anatomical inter-dependencies through a multi-class constraints mechanism. Evaluation on the held-out test set with verified ground truth demonstrates excellent segmentation performance (fovea IoU = 0.812, F1 = 0.894, AED = 4.06 pixels). To demonstrate the efficacy of our synthesis strategy, our GEV-based augmentation technique achieves a detection rate of 98.4% compared to 59.0% for baseline geometric augmentation (paired t-test: t = 8.536, p &lt; 0.001, Cohen\u2019s d = 1.093). Cross-dataset evaluation on REFUGE, MESSIDOR, and ARIA demonstrates competitive localization performance, achieving state-of-the-art Average Euclidean Distance on REFUGE (22.46 \u00b1 18.73 pixels) and MESSIDOR (6.52 \u00b1 5.89 pixels) with robust generalization across diverse imaging protocols. These results establish that explicit anatomical context, rather than mere model complexity, is key to accurate fovea segmentation, offering a robust paradigm for medical image analysis.<\/jats:p>","DOI":"10.1038\/s41598-026-40287-y","type":"journal-article","created":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T21:06:51Z","timestamp":1771967211000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Contextual anatomy-guided deep learning for accurate fovea segmentation in diabetic retinopathy fundus images"],"prefix":"10.1038","volume":"16","author":[{"given":"Sakon","family":"Chankhachon","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Supaporn","family":"Kansomkeat","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Patama","family":"Bhurayanontachai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sathit","family":"Intajag","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,2,24]]},"reference":[{"key":"40287_CR1","doi-asserted-by":"crossref","unstructured":"Prathibha, S. & Siddappaji. Advancing diabetic retinopathy diagnosis with fundus imaging: A comprehensive survey of computer-aided detection, grading and classification methods. Global Transit. 6, 93\u2013112 (2024).","DOI":"10.1016\/j.glt.2024.04.001"},{"key":"40287_CR2","doi-asserted-by":"publisher","first-page":"6618","DOI":"10.1109\/ACCESS.2025.3525499","volume":"13","author":"IS Rajesh","year":"2025","unstructured":"Rajesh, I. S., Reshmi, B. M., Malakreddy, A. B. & Bilakeri, S. Automatic detection of hard exudate in color retinal fundus image and diabetic maculopathy grading. IEEE Access 13, 6618\u20136631 (2025).","journal-title":"IEEE Access"},{"key":"40287_CR3","doi-asserted-by":"crossref","unstructured":"Zheng, S., Pan, L., Chen, J. & Yu, L. Automatic and efficient detection of the fovea center in retinal images. In Proc. 7th Int. Conf. Biomedical Engineering and Informatics (BMEI), 145\u2013150 (2014).","DOI":"10.1109\/BMEI.2014.7002760"},{"key":"40287_CR4","doi-asserted-by":"publisher","first-page":"17753","DOI":"10.1007\/s11042-023-16244-6","volume":"83","author":"J Sigut","year":"2024","unstructured":"Sigut, J., Nu\u00f1ez, O., Fumero, F. & Alay\u00f3n, S. Fovea localization in retinal images using spatial color histograms. Multimed. Tools Appl. 83, 17753\u201317771 (2024).","journal-title":"Multimed. Tools Appl."},{"key":"40287_CR5","doi-asserted-by":"publisher","first-page":"386","DOI":"10.1016\/j.compmedimag.2013.06.002","volume":"37","author":"ME Gegundez-Arias","year":"2013","unstructured":"Gegundez-Arias, M. E., Mar\u00edn, D., Bravo, J. M. & Suero, A. Locating the fovea center position in digital fundus images using thresholding and feature extraction techniques. Comput. Med. Imaging Graph. 37, 386\u2013393 (2013).","journal-title":"Comput. Med. Imaging Graph."},{"key":"40287_CR6","first-page":"1097","volume":"25","author":"A Krizhevsky","year":"2012","unstructured":"Krizhevsky, A., Sutskever, I. & Hinton, G. E. ImageNet classification with deep convolutional neural networks. Adv. Neural. Inf. Process. Syst. 25, 1097\u20131105 (2012).","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"40287_CR7","doi-asserted-by":"publisher","first-page":"6618","DOI":"10.1109\/ACCESS.2025.3525499","volume":"13","author":"IS Rajesh","year":"2025","unstructured":"Rajesh, I. S., Reshmi, B. M., Malakreddy, A. B. & Bilakeri, S. Automatic detection of hard exudate in color retinal fundus image and diabetic maculopathy grading. IEEE Access 13, 6618\u20136631 (2025).","journal-title":"IEEE Access"},{"key":"40287_CR8","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1016\/j.bspc.2017.09.008","volume":"40","author":"B Al-Bander","year":"2018","unstructured":"Al-Bander, B., Al-Nuaimy, W., Williams, B. M. & Zheng, Y. Multiscale sequential convolutional neural networks for simultaneous detection of fovea and optic disc. Biomed. Signal Process. Control 40, 91\u2013101 (2018).","journal-title":"Biomed. Signal Process. Control"},{"key":"40287_CR9","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2020.102001","volume":"111","author":"MK Hasan","year":"2021","unstructured":"Hasan, M. K., Alam, M. A., Elahi, M. T. E., Roy, S. & Mart\u00ed, R. DRNet: Segmentation and localization of optic disc and fovea from diabetic retinopathy image. Artif. Intell. Med. 111, 102001 (2021).","journal-title":"Artif. Intell. Med."},{"key":"40287_CR10","doi-asserted-by":"publisher","first-page":"23107","DOI":"10.1038\/s41598-024-73493-7","volume":"14","author":"R Shalini","year":"2024","unstructured":"Shalini, R. & Gopi, V. P. Multiresolution cascaded attention U-Net for localization and segmentation of optic disc and fovea in fundus images. Sci. Rep. 14, 23107 (2024).","journal-title":"Sci. Rep."},{"key":"40287_CR11","unstructured":"Song, S., Wang, J., Wang, Z., Su, J., Ding, X. & Dang, K. Bilateral-Fuser: A novel multi-cue fusion architecture with anatomical-aware tokens for fovea localization. arXiv:2302.06961 (2023)."},{"key":"40287_CR12","doi-asserted-by":"publisher","first-page":"7217","DOI":"10.1109\/JBHI.2024.3445112","volume":"28","author":"S Song","year":"2024","unstructured":"Song, S. et al. DualStreamFoveaNet: A dual stream fusion architecture with anatomical awareness for robust fovea localization. IEEE J. Biomed. Health Inform. 28, 7217\u20137229 (2024).","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"40287_CR13","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2022.108711","volume":"129","author":"Y Fu","year":"2022","unstructured":"Fu, Y. et al. Fovea localization by blood vessel vector in abnormal fundus images. Pattern Recognit. 129, 108711 (2022).","journal-title":"Pattern Recognit."},{"key":"40287_CR14","doi-asserted-by":"publisher","first-page":"456","DOI":"10.3390\/a7030456","volume":"7","author":"C-Y Yu","year":"2014","unstructured":"Yu, C.-Y., Liu, C.-C. & Yu, S.-S. A fovea localization scheme using vessel origin-based parabolic model. Algorithms 7, 456\u2013470 (2014).","journal-title":"Algorithms"},{"key":"40287_CR15","doi-asserted-by":"crossref","unstructured":"Kumar, J. R. H., Adhikari, R., Kamath, Y., Jampala, R. & Seelamantula, C. S. Automatic delineation of macular regions based on a locally defined contrast function. In Proc. IEEE Int. Conf. Image Processing (ICIP), 1362\u20131366 (2017).","DOI":"10.1109\/ICIP.2017.8296504"},{"key":"40287_CR16","doi-asserted-by":"crossref","unstructured":"Song, S., Dang, K., Yu, Q., Wang, Z., Coenen, F., Su, J. & Ding, X. Bilateral-ViT for robust fovea localization. In Proc. IEEE Int. Symp. Biomedical Imaging (ISBI), 1\u20135 (2022).","DOI":"10.1109\/ISBI52829.2022.9761523"},{"key":"40287_CR17","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.108613","volume":"177","author":"H He","year":"2024","unstructured":"He, H. et al. JOINEDTrans: Prior guided multi-task transformer for joint optic disc\/cup segmentation and fovea detection. Comput. Biol. Med. 177, 108613 (2024).","journal-title":"Comput. Biol. Med."},{"key":"40287_CR18","doi-asserted-by":"crossref","unstructured":"Khumang, S., Kansomkeat, S., Tanomkiat, W. & Intajag, S. Tuberculosis lesion segmentation improvement in X-ray images using contextual background label. IEEE Access 13, 36611\u201336625. https:\/\/doi.org\/10.1109\/ACCESS.2025.3532631 (2025).","DOI":"10.1109\/ACCESS.2025.3532631"},{"key":"40287_CR19","doi-asserted-by":"publisher","first-page":"1885","DOI":"10.1109\/TMI.2023.3242838","volume":"42","author":"Z Li","year":"2023","unstructured":"Li, Z., Kamnitsas, K., Ouyang, C., Chen, C. & Glocker, B. Context label learning: Improving background class representations in semantic segmentation. IEEE Trans. Med. Imaging 42, 1885\u20131896 (2023).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"40287_CR20","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1109\/42.845178","volume":"19","author":"AD Hoover","year":"2000","unstructured":"Hoover, A. D., Kouznetsova, V. & Goldbaum, M. Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response. IEEE Trans. Med. Imaging 19, 203\u2013210 (2000).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"40287_CR21","doi-asserted-by":"publisher","first-page":"501","DOI":"10.1109\/TMI.2004.825627","volume":"23","author":"J Staal","year":"2004","unstructured":"Staal, J., Abr\u00e0moff, M. D., Niemeijer, M., Viergever, M. A. & van Ginneken, B. Ridge-based vessel segmentation in color images of the retina. IEEE Trans. Med. Imaging 23, 501\u2013509 (2004).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"40287_CR22","doi-asserted-by":"publisher","first-page":"25","DOI":"10.3390\/data3030025","volume":"3","author":"P Porwal","year":"2018","unstructured":"Porwal, P., Pachade, S., Kamble, R. & Kokare, M. D. Indian diabetic retinopathy image dataset (IDRiD): A database for diabetic retinopathy screening research. Data 3, 25 (2018).","journal-title":"Data"},{"key":"40287_CR23","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101570","volume":"59","author":"JI Orlando","year":"2020","unstructured":"Orlando, J. I. et al. REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs. Med. Image Anal. 59, 101570 (2020).","journal-title":"Med. Image Anal."},{"key":"40287_CR24","doi-asserted-by":"publisher","first-page":"397","DOI":"10.1038\/s41597-022-01507-y","volume":"9","author":"X Lyu","year":"2022","unstructured":"Lyu, X., Cheng, L. & Zhang, S. The RETA benchmark for retinal vascular tree analysis. Sci. Data 9, 397 (2022).","journal-title":"Sci. Data"},{"key":"40287_CR25","unstructured":"Forrester, J. V., Dick, A. D., McMenamin, P. G., Roberts, F. & Pearlman, E. The Eye: Basic Sciences in Practice (Elsevier Health Sciences, 2015)."},{"key":"40287_CR26","doi-asserted-by":"crossref","unstructured":"Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D. & Batra, D. Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proc. IEEE Int. Conf. Computer Vision (ICCV), 618\u2013626 (2017).","DOI":"10.1109\/ICCV.2017.74"},{"key":"40287_CR27","doi-asserted-by":"crossref","unstructured":"Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F. & Adam, H. Encoder\u2013decoder with atrous separable convolution for semantic image segmentation. In Proc. Eur. Conf. Comput. Vis. (ECCV), Lecture Notes in Computer Science, vol. 11211 (Springer, 2018).","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"40287_CR28","doi-asserted-by":"crossref","unstructured":"Qin, D. et al. MobileNetV4: Universal models for the mobile ecosystem. In Proc. Eur. Conf. Comput. Vis. (ECCV), Lecture Notes in Computer Science, vol. 15098 (Springer, Cham, 2025).","DOI":"10.1007\/978-3-031-73661-2_5"},{"key":"40287_CR29","doi-asserted-by":"publisher","first-page":"596","DOI":"10.1049\/el.2015.3746","volume":"52","author":"S Intajag","year":"2016","unstructured":"Intajag, S., Kansomkeat, S. & Bhurayanontachai, P. Histogram specification with generalised extreme value distribution to enhance retinal images. Electron. Lett. 52, 596\u2013598 (2016).","journal-title":"Electron. Lett."},{"key":"40287_CR30","doi-asserted-by":"publisher","first-page":"748","DOI":"10.1016\/j.jfranklin.2008.04.009","volume":"345","author":"DJJ Farnell","year":"2008","unstructured":"Farnell, D. J. J. et al. Enhancement of blood vessels in digital fundus photographs via the application of multiscale line operators. J. Franklin Inst. 345, 748\u2013765 (2008).","journal-title":"J. Franklin Inst."},{"key":"40287_CR31","doi-asserted-by":"crossref","unstructured":"Vonghirandecha, P., Bhurayanontachai, P., Kansomkeat, S. & Intajag, S. No-reference retinal image sharpness metric using Daubechies wavelet transform. Int. J. Circuits, Syst. Signal Process. 15, 1064\u20131071 (2021).","DOI":"10.46300\/9106.2021.15.115"},{"key":"40287_CR32","doi-asserted-by":"publisher","first-page":"914","DOI":"10.1038\/s41597-024-03739-6","volume":"11","author":"G Lepetit-Aimon","year":"2024","unstructured":"Lepetit-Aimon, G. et al. MAPLES-DR: MESSIDOR anatomical and pathological labels for explainable screening of diabetic retinopathy. Sci. Data 11, 914 (2024).","journal-title":"Sci. Data"},{"key":"40287_CR33","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2020.101939","volume":"60","author":"Y Huang","year":"2020","unstructured":"Huang, Y., Zhong, Z., Yuan, J. & Tang, X. Efficient and robust optic disc detection and fovea localization using region proposal network and cascaded network. Biomed. Signal Process. Control 60, 101939 (2020).","journal-title":"Biomed. Signal Process. Control"},{"key":"40287_CR34","doi-asserted-by":"crossref","unstructured":"Meyer, M. I., Galdran, A., Mendon\u00e7a, A. M. & Campilho, A. A pixel-wise distance regression approach for joint retinal optical disc and fovea detection. In Proc. Medical Image Computing and Computer Assisted Intervention (MICCAI), 39\u201347 (2018).","DOI":"10.1007\/978-3-030-00934-2_5"},{"key":"40287_CR35","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1109\/TMI.2020.3023254","volume":"40","author":"R Xie","year":"2021","unstructured":"Xie, R. et al. End-to-end fovea localisation in colour fundus images with a hierarchical deep regression network. IEEE Trans. Med. Imaging 40, 116\u2013128 (2021).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"40287_CR36","first-page":"09860","volume":"2110","author":"S Song","year":"2022","unstructured":"Song, S. et al. Bilateral-ViT for robust fovea localization. arXiv 2110, 09860 (2022).","journal-title":"Bilateral-ViT for robust fovea localization. arXiv"},{"key":"40287_CR37","doi-asserted-by":"publisher","first-page":"159071","DOI":"10.1109\/ACCESS.2021.3127280","volume":"9","author":"BJ Bhatkalkar","year":"2021","unstructured":"Bhatkalkar, B. J., Nayak, S. V., Shenoy, S. V. & Arjunan, R. V. FundusPosNet: A deep learning driven heatmap regression model for the joint localization of optic disc and fovea centers in color fundus images. IEEE Access 9, 159071\u2013159080 (2021).","journal-title":"IEEE Access"},{"key":"40287_CR38","doi-asserted-by":"publisher","first-page":"2315","DOI":"10.1109\/JBHI.2020.2971593","volume":"24","author":"X Guo","year":"2020","unstructured":"Guo, X. et al. Robust fovea localization based on symmetry measure. IEEE J. Biomed. Health Inform. 24, 2315\u20132326 (2020).","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"40287_CR39","unstructured":"He, H., Lin, L., Cai, Z. & Tang, X. JOINED: Prior guided multi-task learning for joint optic disc\/cup segmentation and fovea detection. arXiv:2203.00461 (2022)."},{"key":"40287_CR40","unstructured":"Babu, S. C., Maiya, S. R. & Elango, S. Relation networks for optic disc and fovea localization in retinal images. arXiv:1812.00883 (2018)."},{"key":"40287_CR41","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2025.110125","volume":"191","author":"JP Medhi","year":"2025","unstructured":"Medhi, J. P., Nirmala, S. R., Borah, K., Bhattacharjee, D. & Dandapat, S. Sharper insights: Adaptive ellipse-template for robust fovea localization in challenging retinal landscapes. Comput. Biol. Med. 191, 110125 (2025).","journal-title":"Comput. Biol. Med."},{"key":"40287_CR42","doi-asserted-by":"publisher","first-page":"1631","DOI":"10.3390\/sym15091631","volume":"15","author":"HA Wibawa","year":"2023","unstructured":"Wibawa, H. A., Sumiharto, R., Harjoko, A. & Sasongko, M. B. The methods of determining temporal direction based on asymmetric information of the optic disc for optimal fovea detection. Symmetry 15, 1631 (2023).","journal-title":"Symmetry"},{"key":"40287_CR43","doi-asserted-by":"crossref","unstructured":"Hochberg, Y. & Tamhane, A. C. Multiple Comparison Procedures (John Wiley & Sons, 1987).","DOI":"10.1002\/9780470316672"},{"key":"40287_CR44","doi-asserted-by":"publisher","first-page":"231","DOI":"10.5566\/ias.1155","volume":"33","author":"E Decenci\u00e8re","year":"2014","unstructured":"Decenci\u00e8re, E. et al. Feedback on a publicly distributed database: the MESSIDOR database. Image Anal. Stereol. 33, 231\u2013234 (2014).","journal-title":"Image Anal. Stereol."},{"key":"40287_CR45","doi-asserted-by":"publisher","first-page":"1463233","DOI":"10.3389\/frai.2025.1463233","volume":"8","author":"W Li","year":"2025","unstructured":"Li, W., Cao, L. & Deng, H. FLA-UNet: feature-location attention U-Net for foveal avascular zone segmentation in OCTA images. Front. Artif. Intell. 8, 1463233 (2025).","journal-title":"Front. Artif. Intell."}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-40287-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-40287-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-40287-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T10:18:57Z","timestamp":1774606737000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-40287-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,24]]},"references-count":45,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["40287"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-40287-y","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,24]]},"assertion":[{"value":"17 October 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 February 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 February 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"10388"}}