{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T19:00:17Z","timestamp":1772650817399,"version":"3.50.1"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T00:00:00Z","timestamp":1752710400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Since optical coherence tomography angiography (OCTA) is non-invasive and non-contact, it is widely used in the study of retinal disease detection. As a key indicator for retinal disease detection, accurate segmentation of foveal avascular zone (FAZ) has an important impact on clinical application. Although the U-Net and its existing improvement methods have achieved good performance on FAZ segmentation, their generalization ability and segmentation accuracy can be further improved by exploring more effective improvement strategies.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We propose a novel improved method named Feature-location Attention U-Net (FLA-UNet) by introducing new designed feature-location attention blocks (FLABs) into U-Net and using a joint loss function. The FLAB consists of feature-aware blocks and location-aware blocks in parallel, and is embed into each decoder of U-Net to integrate more marginal information of FAZ and strengthen the connection between target region and boundary information. The joint loss function is composed of the cross-entropy loss (CE loss) function and the Dice coefficient loss (Dice loss) function, and by adjusting the weights of them, the performance of the network on boundary and internal segmentation can be comprehensively considered to improve its accuracy and robustness for FAZ segmentation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The qualitative and quantitative comparative experiments on the three datasets of OCTAGON, FAZID and OCTA-500 show that, our proposed FLA-UNet achieves better segmentation quality, and is superior to other existing state-of-the-art methods in terms of the MIoU, ACC and Dice coefficient.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>The proposed FLA-UNet can effectively improve the accuracy and robustness of FAZ segmentation in OCTA images by introducing feature-location attention blocks into U-Net and using a joint loss function. This has laid a solid theoretical foundation for its application in auxiliary diagnosis of fundus diseases.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1463233","type":"journal-article","created":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T05:24:12Z","timestamp":1752729852000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["FLA-UNet: feature-location attention U-Net for foveal avascular zone segmentation in OCTA images"],"prefix":"10.3389","volume":"8","author":[{"given":"Wei","family":"Li","sequence":"first","affiliation":[]},{"given":"Li","family":"Cao","sequence":"additional","affiliation":[]},{"given":"He","family":"Deng","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,7,17]]},"reference":[{"key":"ref1","doi-asserted-by":"crossref","DOI":"10.1117\/12.2567580","article-title":"The foveal avascular zone image database (FAZID)","author":"Agarwal","year":"2020"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"872","DOI":"10.3390\/jcm8060872","article-title":"Supervised machine learning based multi-task artificial intelligence classification of retinopathies","volume":"8","author":"Alam","year":"2019","journal-title":"J. Clin. Med."},{"key":"ref3","doi-asserted-by":"crossref","DOI":"10.1117\/12.2544817","article-title":"Comparison of foveal avascular zone in diabetic retinopathy, high myopia, and normal fundus images","author":"Balaji","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1016\/j.dsp.2009.06.005","article-title":"Detection of the foveal avascular zone on retinal angiograms using Markov random fields","volume":"20","author":"Bourennane","year":"2010","journal-title":"Digit. Signal Process."},{"key":"ref5","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP43922.2022.9747503","article-title":"FAZ-BV: a diabetic macular ischemia grading framework combining Faz attention network and blood vessel enhancement filters","author":"Chen","year":"2022"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"602","DOI":"10.1097\/OPX.0b013e3182504227","article-title":"Foveal avascular zone and its relationship to foveal pit shape","volume":"89","author":"Chui","year":"2012","journal-title":"Optom. Vis. Sci."},{"key":"ref7","first-page":"505","article-title":"Automatic wide field registration and mosaicking of OCTA images using vascularity information","author":"D\u00edaz","year":""},{"key":"ref8","doi-asserted-by":"publisher","first-page":"e0212364","DOI":"10.1371\/journal.pone.0212364","article-title":"Automatic segmentation of the foveal avascular zone in ophthalmological OCT-A images","volume":"14","author":"D\u00edaz","year":"","journal-title":"PLoS One"},{"key":"ref9","first-page":"70","article-title":"Automated segmentation of the foveal avascular zone in angiograms using region growing","volume":"37","author":"Ghassemi","year":"2007","journal-title":"Comput. Biol. Med."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"2281","DOI":"10.1109\/TMI.2019.2903562","article-title":"CE-net: context encoder network for 2D medical image segmentation","volume":"38","author":"Gu","year":"2019","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"5147","DOI":"10.1364\/BOE.9.005147","article-title":"Med-net, a neural network for automated detection of avascular area in OCT angiography","volume":"9","author":"Guo","year":"2018","journal-title":"Biomed. Opt. Express"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2021.102456","article-title":"Can deep learning improve the automatic segmentation of deep foveal avascular zone in optical coherence tomography angiography?","volume":"66","author":"Guo","year":"2021","journal-title":"Biomed. Signal Process. Control"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1186\/s42492-019-0031-8","article-title":"Automatic quantification of superficial foveal avascular zone in optical coherence tomography angiography implemented with deep learning","volume":"2","author":"Guo","year":"2019","journal-title":"Vis. Comput. Ind. Biomed. Art"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2022.3193188","article-title":"Joint-seg: treat foveal avascular zone and retinal vessel segmentation in OCTA images as a joint task","volume":"71","author":"Hu","year":"2022","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref15","article-title":"U-net 3+: a full-scale connected U-net for medical image segmentation","author":"Huang","year":"2020"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"3393","DOI":"10.32604\/cmc.2023.040091","article-title":"DT-net: joint dual-input transformer and CNN for retinal vessel segmentation","volume":"76","author":"Jia","year":"2023","journal-title":"Comput. Mater. Contin."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1016\/j.preteyeres.2017.07.002","article-title":"Optical coherence tomography angiography: a comprehensive review of current methods and clinical applications","volume":"60","author":"Kashani","year":"2017","journal-title":"Prog. Retin. Eye Res."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"3343","DOI":"10.1109\/TMI.2020.2992244","article-title":"Image projection network: 3D to 2D image segmentation in octa images","volume":"39","author":"Li","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"103092","DOI":"10.1016\/j.media.2024.103092","article-title":"OCTA-500: a retinal dataset for optical coherence tomography angiography study","volume":"93","author":"Li","year":"2020","journal-title":"Med. Image Anal."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"3830","DOI":"10.1002\/mp.15608","article-title":"RPS-net: an effective retinal image projection segmentation network for retinal vessels and foveal avascular zone based on OCTA data","volume":"49","author":"Li","year":"2022","journal-title":"Med. Phys."},{"key":"ref21","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-87237-3_7","article-title":"BSDA-Net: a boundary shape and distance aware joint learning framework for segmenting and classifying OCTA images","author":"Lin","year":"2021"},{"key":"ref22","first-page":"1","article-title":"Automated segmentation of foveal avascular zone in optical coherence tomography angiography images using sparse deep learning","volume":"24","author":"Liu","year":"2019","journal-title":"J. Biomed. Opt."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"6535","DOI":"10.1038\/s41598-025-90709-6","article-title":"Dual-branch image projection network for geographic atrophy segmentation in retinal OCT images","volume":"15","author":"Liu","year":"2025","journal-title":"Sci. Rep."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1142\/S1793545822420019","article-title":"Automatic segmentation of foveal avascular zone based on adaptive watershed algorithm in retinal optical coherence tomography angiography images","volume":"15","author":"Liu","year":"2022","journal-title":"J. Innov. Opt. Health Sci."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"101874","DOI":"10.1016\/j.media.2020.101874","article-title":"CS2-Net: deep learning segmentation of curvilinear structures in medical imaging","volume":"67","author":"Mou","year":"2021","journal-title":"Med. Image Anal."},{"key":"ref26","first-page":"234","article-title":"U-net: Convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1385742","DOI":"10.3389\/fradi.2024.1385742","article-title":"A systematic literature review: deep learning techniques for synthetic medical image generation and their applications in radiotherapy","volume":"4","author":"Sherwani","year":"2024","journal-title":"Front. Radiol."},{"key":"ref28","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-64583-0_50","article-title":"Evaluating the impact of training loss on MR to synthetic CT conversion","author":"Sherwani","year":"2020"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"795","DOI":"10.1016\/S0031-3203(00)00032-7","article-title":"Segmentation of macular fluorescein angiographies. A statistical approach","volume":"34","author":"Sim\u00f3","year":"2001","journal-title":"Pattern Recogn."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"112704","DOI":"10.1016\/j.knosys.2024.112704","article-title":"ODDF-net: multi-object segmentation in 3D retinal OCTA using optical density and disease features","volume":"306","author":"Yang","year":"2024","journal-title":"Knowl.-Based Syst."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"104249","DOI":"10.1016\/j.bspc.2022.104249","article-title":"Multi-task thyroid tumor segmentation based on the joint loss function","volume":"79","author":"Yang","year":"2023","journal-title":"Biomed. Signal process. Control"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"873","DOI":"10.1007\/s00417-015-3143-7","article-title":"A morphological study of the foveal avascular zone in patients with diabetes mellitus using optical coherence tomography angiography","volume":"254","author":"Yang","year":"2016","journal-title":"Graefes Arch. Clin. Exp. Ophthalmol."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1856","DOI":"10.1109\/TMI.2019.2959609","article-title":"U-net++: redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Zhou","year":"2020","journal-title":"IEEE Trans. Med. Imaging"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1463233\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T05:24:15Z","timestamp":1752729855000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1463233\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,17]]},"references-count":33,"alternative-id":["10.3389\/frai.2025.1463233"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1463233","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,17]]},"article-number":"1463233"}}