{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T19:15:03Z","timestamp":1792782903217,"version":"4.5.0"},"reference-count":30,"publisher":"Springer Science and Business Media LLC","issue":"3","license":[{"start":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T00:00:00Z","timestamp":1786406400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T00:00:00Z","timestamp":1786406400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"INESC TEC - Instituto de Engenharia de Sistemas e Computadores, Tecnologia e Ci\u00eancia","id":[{"id":"https:\/\/ror.org\/033wn8m60","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Pattern Anal Applic"],"published-print":{"date-parts":[[2026,9]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The segmentation of the foveal avascular zone (FAZ) in optical coherence tomography angiography (OCTA) images plays a crucial role in diagnosing and monitoring ocular diseases such as diabetic retinopathy (DR) and age-related macular degeneration (AMD). However, accurate FAZ segmentation remains challenging due to image quality and variability. This paper provides a comprehensive review of FAZ segmentation techniques, including traditional image processing methods and recent deep learning-based approaches. We propose two novel deep learning methodologies: a multitask learning framework that integrates vessel and FAZ segmentation, and a conditionally trained network that employs vessel-aware loss functions. For this purpose, we did a preliminary study of blood vessel segmentation backbones, in which OCTA-Net and COSNet, with and without several architectural and training modifications, were extensively evaluated on multiple OCTA datasets. The outcome of this study was then used for selecting the vessel backbone adopted in the proposed approaches for FAZ segmentation. The performance of these methods was evaluated on the OCTA-500 dataset using the Dice coefficient, Jaccard index, 95% Hausdorff distance, and average symmetric surface distance. Experimental results demonstrate that the multitask segmentation framework outperforms existing state-of-the-art methods, achieving superior FAZ boundary delineation and segmentation accuracy. The conditionally trained network also improves upon standard U-Net-based approaches but exhibits limitations in refining the FAZ contours.<\/jats:p>","DOI":"10.1007\/s10044-026-01699-8","type":"journal-article","created":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T21:25:12Z","timestamp":1786483512000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multitask learning approach for foveal avascular zone segmentation in OCTA images"],"prefix":"10.1007","volume":"29","author":[{"given":"T\u00e2nia","family":"Melo","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matilde","family":"Costa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u00c2ngela","family":"Carneiro","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aur\u00e9lio","family":"Campilho","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ana Maria","family":"Mendon\u00e7a","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,8,11]]},"reference":[{"issue":"2","key":"1699_CR1","doi-asserted-by":"publisher","first-page":"1056","DOI":"10.1364\/BOE.8.001056","volume":"8","author":"C-L Chen","year":"2017","unstructured":"Chen C-L, Wang RK (2017) Optical coherence tomography based angiography. Biomed Opt Express 8(2):1056\u20131082. https:\/\/doi.org\/10.1364\/BOE.8.001056","journal-title":"Biomed Opt Express"},{"issue":"8","key":"1699_CR2","first-page":"1125","volume":"64","author":"AC Benech","year":"2023","unstructured":"Benech AC, Wang Z, Bal AB et al (2023) Deep learning segmentation of foveal avascular zone (FAZ) in optical coherence tomography angiography (OCTA) of nonproliferative diabetic retinopathy. Investigative Ophthalmol Visual Sci 64(8):1125","journal-title":"Investigative Ophthalmol Visual Sci"},{"issue":"12","key":"1699_CR3","doi-asserted-by":"publisher","first-page":"3969","DOI":"10.1109\/TMI.2022.3202183","volume":"41","author":"J Hao","year":"2022","unstructured":"Hao J, Shen T, Zhu X et al (2022) Retinal structure detection in OCTA image via voting-based multitask learning. IEEE Trans Med Imaging 41(12):3969\u20133980. https:\/\/doi.org\/10.1109\/TMI.2022.3202183","journal-title":"IEEE Trans Med Imaging"},{"key":"1699_CR4","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2025.112790","volume":"173","author":"Z Zhan","year":"2026","unstructured":"Zhan Z, Zhang X, Zhao M, Zhang Y, Zhou W (2026) BLPNet: Boosting localization perception network for foveal avascular zone segmentation. Pattern Recogn 173:112790. https:\/\/doi.org\/10.1016\/j.patcog.2025.112790","journal-title":"Pattern Recogn"},{"key":"1699_CR5","doi-asserted-by":"publisher","unstructured":"Peng L, Lin L, Cheng P et al. (2021) FARGO: A joint framework for FAZ and RV segmentation from OCTA images. In: Fu, H., Garvin, M.K., MacGillivray, T., et al. (eds.) Ophthalmic Medical Image Analysis, pp. 42\u201351. Springer, Berlin, Heidelberg . https:\/\/doi.org\/10.1007\/978-3-030-87000-3_5","DOI":"10.1007\/978-3-030-87000-3_5"},{"key":"1699_CR6","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103092","volume":"93","author":"M Li","year":"2024","unstructured":"Li M, Huang K, Xu Q, Yang J, Zhang Y, Ji Z, Xie K, Yuan S, Liu Q, Chen Q (2024) OCTA-500: A retinal dataset for optical coherence tomography angiography study. Med Image Anal 93:103092. https:\/\/doi.org\/10.1016\/j.media.2024.103092","journal-title":"Med Image Anal"},{"issue":"6","key":"1699_CR7","doi-asserted-by":"publisher","first-page":"2212","DOI":"10.1167\/iovs.17-23498","volume":"59","author":"Y Lu","year":"2018","unstructured":"Lu Y, Simonett JM, Wang J et al (2018) Evaluation of automatically quantified foveal avascular zone metrics for diagnosis of diabetic retinopathy using optical coherence tomography angiography. Investigative Ophthalmol Vis Sci 59(6):2212\u20132221. https:\/\/doi.org\/10.1167\/iovs.17-23498","journal-title":"Investigative Ophthalmol Vis Sci"},{"issue":"4","key":"1699_CR8","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1001\/jamaophthalmol.2015.5658","volume":"134","author":"TS Hwang","year":"2016","unstructured":"Hwang TS, Gao SS, Liu L et al (2016) Automated quantification of capillary nonperfusion using optical coherence tomography angiography in diabetic retinopathy. JAMA Ophthalmol 134(4):367\u2013373. https:\/\/doi.org\/10.1001\/jamaophthalmol.2015.5658","journal-title":"JAMA Ophthalmol"},{"issue":"2","key":"1699_CR9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1371\/journal.pone.0212364","volume":"14","author":"M D\u00edaz","year":"2019","unstructured":"D\u00edaz M, Novo J, Cutr\u00edn P et al (2019) Automatic segmentation of the foveal avascular zone in ophthalmological OCT-A images. PLoS ONE 14(2):1\u201322. https:\/\/doi.org\/10.1371\/journal.pone.0212364","journal-title":"PLoS ONE"},{"issue":"1","key":"1699_CR10","doi-asserted-by":"publisher","first-page":"2242001","DOI":"10.1142\/S1793545822420019","volume":"15","author":"J Liu","year":"2022","unstructured":"Liu J, Yan S, Lu N et al (2022) Automatic segmentation of foveal avascular zone based on adaptive watershed algorithm in retinal optical coherence tomography angiography images. J Innovative Opt Health Sci 15(1):2242001. https:\/\/doi.org\/10.1142\/S1793545822420019","journal-title":"J Innovative Opt Health Sci"},{"key":"1699_CR11","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2021.102456","volume":"66","author":"M Guo","year":"2021","unstructured":"Guo M, Zhao M, Cheong AM et al (2021) Can deep learning improve the automatic segmentation of deep foveal avascular zone in optical coherence tomography angiography? Biomed Signal Process Control 66:102456. https:\/\/doi.org\/10.1016\/j.bspc.2021.102456","journal-title":"Biomed Signal Process Control"},{"key":"1699_CR12","doi-asserted-by":"publisher","unstructured":"Meena G, Mohbey K, Acharya M, Lokesh K (2023) An improved convolutional neural network-based model for detecting brain tumors from augmented MRI images. Journal of Autonomous Intelligence 6. https:\/\/doi.org\/10.32629\/jai.v6i1.561","DOI":"10.32629\/jai.v6i1.561"},{"key":"1699_CR13","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2025.102995","volume":"118","author":"X Li","year":"2025","unstructured":"Li X, Li L, Jiang Y, Wang H, Qiao X, Feng T, Luo H, Zhao Y (2025) Vision-language models in medical image analysis: From simple fusion to general large models. Inform Fus 118:102995. https:\/\/doi.org\/10.1016\/j.inffus.2025.102995","journal-title":"Inform Fus"},{"key":"1699_CR14","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2025.103819","volume":"107","author":"X Li","year":"2026","unstructured":"Li X, Li L, Li M, Yan P, Feng T, Luo H, Zhao Y, Yin S (2026) Knowledge distillation and teacher-student learning in medical imaging: Comprehensive overview, pivotal role, and future directions. Med Image Anal 107:103819. https:\/\/doi.org\/10.1016\/j.media.2025.103819","journal-title":"Med Image Anal"},{"key":"1699_CR15","doi-asserted-by":"publisher","unstructured":"Tan X, Chen X, Meng Q, Shi F, Xiang D, Chen Z, Pan L, Zhu W (2023) OCT$$^2$$Former: A retinal OCT-angiography vessel segmentation transformer. Comput Methods Programs Biomed 233:107454. https:\/\/doi.org\/10.1016\/j.cmpb.2023.107454","DOI":"10.1016\/j.cmpb.2023.107454"},{"key":"1699_CR16","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2023.104604","volume":"83","author":"X Liu","year":"2023","unstructured":"Liu X, Zhang D, Yao J, Tang J (2023) Transformer and convolutional based dual branch network for retinal vessel segmentation in OCTA images. Biomed Signal Process Control 83:104604. https:\/\/doi.org\/10.1016\/j.bspc.2023.104604","journal-title":"Biomed Signal Process Control"},{"issue":"1","key":"1699_CR17","doi-asserted-by":"publisher","first-page":"24757","DOI":"10.1038\/s41598-025-07448-x","volume":"15","author":"F Sedighin","year":"2025","unstructured":"Sedighin F, Rezaei F, Monemian M (2025) Low rank based FAZ segmentation in OCTA images. Sci Rep 15(1):24757. https:\/\/doi.org\/10.1038\/s41598-025-07448-x","journal-title":"Sci Rep"},{"issue":"1","key":"1699_CR18","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.imed.2025.05.008","volume":"6","author":"K Ma","year":"2026","unstructured":"Ma K, Zhang J, Hu X, Chen Y, Wang G, Huang Y, Yuan J, Xiao P (2026) Deep learning-based segmentation and quantitative analysis of retinal microstructures in optical coherence tomography angiography images using RS_Unet3+. Intell Med 6(1):12\u201323. https:\/\/doi.org\/10.1016\/j.imed.2025.05.008","journal-title":"Intell Med"},{"key":"1699_CR19","doi-asserted-by":"publisher","unstructured":"Li W, Cao L, Deng H (2025) FLA-UNet: feature-location attention U-Net for foveal avascular zone segmentation in OCTA images. Frontiers in Artificial Intelligence 8. https:\/\/doi.org\/10.3389\/frai.2025.1463233","DOI":"10.3389\/frai.2025.1463233"},{"issue":"1","key":"1699_CR20","doi-asserted-by":"publisher","first-page":"15368","DOI":"10.1038\/s41598-020-72358-z","volume":"10","author":"N Mehta","year":"2020","unstructured":"Mehta N, Braun PX, Gendelman I et al (2020) Repeatability of binarization thresholding methods for optical coherence tomography angiography image quantification. Sci Rep 10(1):15368. https:\/\/doi.org\/10.1038\/s41598-020-72358-z","journal-title":"Sci Rep"},{"issue":"6","key":"1699_CR21","doi-asserted-by":"publisher","first-page":"66008","DOI":"10.1117\/1.JBO.21.6.066008","volume":"21","author":"Z Chu","year":"2016","unstructured":"Chu Z, Lin J, Gao C et al (2016) Quantitative assessment of the retinal microvasculature using optical coherence tomography angiography. J Biomed Opt 21(6):66008. https:\/\/doi.org\/10.1117\/1.JBO.21.6.066008","journal-title":"J Biomed Opt"},{"key":"1699_CR22","doi-asserted-by":"publisher","unstructured":"Jin J, Yang L, Zhang X et al (2013) Vascular tree segmentation in medical images using hessian-based multiscale filtering and level set method. Mathematical Methods and Applications in Medical Imaging 502013. https:\/\/doi.org\/10.1155\/2013\/502013","DOI":"10.1155\/2013\/502013"},{"issue":"12","key":"1699_CR23","doi-asserted-by":"publisher","first-page":"5604","DOI":"10.1364\/BOE.8.005604","volume":"8","author":"A Li","year":"2017","unstructured":"Li A, You J, Du C et al (2017) Automated segmentation and quantification of OCT angiography for tracking angiogenesis progression. Biomed Opt Express 8(12):5604\u20135616. https:\/\/doi.org\/10.1364\/BOE.8.005604","journal-title":"Biomed Opt Express"},{"issue":"7","key":"1699_CR24","doi-asserted-by":"publisher","first-page":"1010","DOI":"10.1109\/83.931095","volume":"10","author":"F Zana","year":"2001","unstructured":"Zana F, Klein J-C (2001) Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation. IEEE Trans Image Process 10(7):1010\u20131019. https:\/\/doi.org\/10.1109\/83.931095","journal-title":"IEEE Trans Image Process"},{"key":"1699_CR25","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101874","volume":"67","author":"L Mou","year":"2021","unstructured":"Mou L, Zhao Y, Fu H et al (2021) CS2-Net: Deep learning segmentation of curvilinear structures in medical imaging. Med Image Anal 67:101874. https:\/\/doi.org\/10.1016\/j.media.2020.101874","journal-title":"Med Image Anal"},{"issue":"13","key":"1699_CR26","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1167\/tvst.9.13.5","volume":"9","author":"Y Giarratano","year":"2020","unstructured":"Giarratano Y, Bianchi E, Gray C et al (2020) Automated segmentation of optical coherence tomography angiography images: Benchmark data and clinically relevant metrics. Trans Vis Sci Technol 9(13):5. https:\/\/doi.org\/10.1167\/tvst.9.13.5","journal-title":"Trans Vis Sci Technol"},{"key":"1699_CR27","doi-asserted-by":"publisher","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., et al. (eds.) Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015, pp. 234\u2013241. Springer, Cham. https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28","DOI":"10.1007\/978-3-319-24574-4_28"},{"issue":"3","key":"1699_CR28","doi-asserted-by":"publisher","first-page":"928","DOI":"10.1109\/TMI.2020.3042802","volume":"40","author":"Y Ma","year":"2021","unstructured":"Ma Y, Hao H, Xie J et al (2021) ROSE: A retinal OCT-angiography vessel segmentation dataset and new model. IEEE Trans Med Imaging 40(3):928\u2013939. https:\/\/doi.org\/10.1109\/TMI.2020.3042802","journal-title":"IEEE Trans Med Imaging"},{"issue":"24","key":"1699_CR29","doi-asserted-by":"publisher","first-page":"9847","DOI":"10.3390\/s22249847","volume":"22","author":"Z Ma","year":"2022","unstructured":"Ma Z, Feng D, Wang J et al (2022) Retinal OCTA image segmentation based on global contrastive learning. Sensors 22(24):9847. https:\/\/doi.org\/10.3390\/s22249847","journal-title":"Sensors"},{"key":"1699_CR30","unstructured":"Costa MC (2023) Segmentation and Characterization of the Vascular Network in OCTA Images. Master\u2019s dissertation, Faculty of Engineering - University of Porto"}],"container-title":["Pattern Analysis and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10044-026-01699-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10044-026-01699-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10044-026-01699-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T11:19:39Z","timestamp":1790075979000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10044-026-01699-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,11]]},"references-count":30,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,9]]}},"alternative-id":["1699"],"URL":"https:\/\/doi.org\/10.1007\/s10044-026-01699-8","relation":{},"ISSN":["1433-7541","1433-755X"],"issn-type":[{"value":"1433-7541","type":"print"},{"value":"1433-755X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,11]]},"assertion":[{"value":"30 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 May 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 August 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no conflict of interest.","order":1,"name":"Ethics","label":"Conflict of interest","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":3,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"165"}}