{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T07:47:54Z","timestamp":1789631274556,"version":"build-2803163510"},"reference-count":42,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T00:00:00Z","timestamp":1772668800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Central Government-Guided Local Science and Technology Development Fund","award":["YDZJSX2025D081"],"award-info":[{"award-number":["YDZJSX2025D081"]}]},{"name":"Support for Young Top Talents of \u201cSanjin Talents\u201d","award":["SJYC2024455"],"award-info":[{"award-number":["SJYC2024455"]}]},{"name":"Fund Program for the Scientific Activities of Selected Returned Overseas Professionals in Shanxi Province","award":["20250064"],"award-info":[{"award-number":["20250064"]}]},{"name":"Doctoral Fund of Shanxi Eye Hospital","award":["B202404"],"award-info":[{"award-number":["B202404"]}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Currently, publicly available Optical Coherence Tomography (OCT) datasets are commonly plagued by limited coverage of disease categories, scarce samples and severe class imbalance, which leads to insufficient generalization ability of deep learning models in real-world clinical settings. This study aims to construct a high-quality OCT dataset encompassing six key types of fundus lesions and normal controls, and to systematically evaluate the improvement effect of training strategies for multi-source data fusion on the performance of multi-class classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We integrated local clinical data from Shanxi Eye Hospital with the latest public dataset OCTDL to establish a combined dataset. This dataset consists of 6,165 images, covering seven categories: age-related macular degeneration (AMD), diabetic macular edema (DME), retinal artery occlusion (RAO), retinal vein occlusion (RVO), epiretinal membrane (ERM), vitreomacular interface disease (VID), and normal controls (NO). On this basis, six representative deep learning architectures were selected, and two training paradigms were compared under unified experimental settings: (1) Training exclusively on open-source OCTDL data (S1); (2) Joint training using both local data and OCTDL data (S2). All models were evaluated on the identical OCTDL test set. A comprehensive analysis was conducted using multi-dimensional metrics including accuracy, weighted F1-score, class-specific recall, and area under the curve (AUC), with a particular focus on the misdiagnosis rate.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The S1 strategy exhibited significantly limited model recognition capability due to the extremely small sample sizes of certain categories. In contrast, the S2 strategy markedly improved the overall performance of the models. Confusion matrix analysis demonstrated that ViT-Base achieved the optimal performance under the S2 strategy: the accuracy reached 93.61%, the misdiagnosis rate of RAO was reduced to 0%, the misdiagnosis rate of AMD was controlled at 1.34%, and the misdiagnosis rate of RVO decreased from 14.89 to 8.51%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Multi-source data fusion serves as an effective approach to enhance the robustness of OCT multi-category classification models, and it can notably strengthen the recognition capability for certain diseases in particular. This study not only verifies the universal benefits of this strategy but also reveals the critical impact of model selection on the transfer learning effect.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1775911","type":"journal-article","created":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T05:25:56Z","timestamp":1772688356000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Evaluation of classification performance for six types of fundus diseases in OCT images based on multi-source training strategy"],"prefix":"10.3389","volume":"13","author":[{"given":"Biao","family":"Guo","sequence":"first","affiliation":[{"name":"Netchina Huaxin Technology Co., Ltd.","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daqing","family":"Wang","sequence":"additional","affiliation":[{"name":"Netchina Huaxin Technology Co., Ltd.","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhuo","family":"Zhao","sequence":"additional","affiliation":[{"name":"Shanxi Eye Hospital","place":["Taiyuan, Shanxi, China"]},{"name":"Shanxi Medical University","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenchao","family":"Liu","sequence":"additional","affiliation":[{"name":"Shanxi Eye Hospital","place":["Taiyuan, Shanxi, China"]},{"name":"Shanxi Medical University","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jia","family":"Hou","sequence":"additional","affiliation":[{"name":"Shanxi Eye Hospital","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruilin","family":"Liang","sequence":"additional","affiliation":[{"name":"The 985th Hospital of the Joint Logistics Support Force of the Chinese People's Liberation Army","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lijuan","family":"Zhang","sequence":"additional","affiliation":[{"name":"Shanxi Eye Hospital","place":["Taiyuan, Shanxi, China"]},{"name":"Shanxi Medical University","place":["Taiyuan, Shanxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,3,5]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/j.cmpb.2017.01.007","article-title":"An improved arteriovenous classification method for the early diagnostics of various diseases in retinal image","volume":"141","author":"Xu","year":"2017","journal-title":"Comput Methods Prog Biomed"},{"key":"ref2","first-page":"465","author":"Najeeb","year":"2018"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"3360","DOI":"10.1109\/ACCESS.2018.2888639","article-title":"Diagnosis of diabetic retinopathy using deep neural networks","volume":"7","author":"Gao","year":"2019","journal-title":"IEEE Access"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1959","DOI":"10.1109\/TMI.2019.2898414","article-title":"Attention to lesion: lesion - aware convolutional neural network for retinal optical coherence tomography image classification","volume":"38","author":"Fang","year":"2019","journal-title":"IEEE Trans Med Imaging"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1016\/j.cmpb.2019.06.016","article-title":"Classification of diabetes - related retinal diseases using a deep learning approach in optical coherence tomography","volume":"178","author":"Perdomo","year":"2019","journal-title":"Comput Methods Prog Biomed"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1007\/s11517-019-02066-y","article-title":"AOCT-NET: a convolutional network automated classification of multiclass retinal diseases using spectral - domain optical coherence tomography images","volume":"58","author":"Alqudah","year":"2020","journal-title":"Med Biol Eng Comput"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"394","DOI":"10.1002\/jemt.23596","article-title":"Classification of retinal images based on convolutional neural network","volume":"84","author":"El\u2010Hag","year":"2021","journal-title":"Microsc Res Tech"},{"key":"ref8","first-page":"1","author":"Kim","year":"2021"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1007\/s11517-021-02321-1","article-title":"Feasibility study to improve deep learning in OCT diagnosis of rare retinal diseases with few - shot classification","volume":"59","author":"Yoo","year":"2021","journal-title":"Med Biol Eng Comput"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1167\/tvst.11.10.39","article-title":"Deep ensemble learning for retinal image classification","volume":"11","author":"Ho","year":"2022","journal-title":"Transl Vis Sci Technol"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"e0261285","DOI":"10.1371\/journal.pone.0261285","article-title":"MHANet: A hybrid attention mechanism for retinal diseases classification","volume":"16","author":"Xu","year":"2021","journal-title":"PLoS One"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"105877","DOI":"10.1016\/j.cmpb.2020.105877","article-title":"OctNET: A lightweight CNN for retinal disease classification from optical coherence tomography images","volume":"200","author":"Sunija","year":"2021","journal-title":"Comput Methods Programs Biomed"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"e112","DOI":"10.47750\/jptcp.2022.945","article-title":"Comparison of the proposed DCNN model with standard CNN architectures for retinal diseases classification","volume":"29","author":"Mohan","year":"2022","journal-title":"J Popul Ther Clin Pharmacol"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"3858","DOI":"10.1109\/EMBC48229.2022.9871322","article-title":"Automatic classification of macular diseases from OCT images using CNN guided with edge convolutional layer","volume":"2022","author":"Esfahani","year":"2022","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"906","DOI":"10.1007\/s12539-022-00533-z","article-title":"Deep residual network for diagnosis of retinal diseases using optical coherence tomography images","volume":"14","author":"Asif","year":"2022","journal-title":"Interdiscip Sci"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"102608","DOI":"10.1016\/j.media.2022.102608","article-title":"Focused attention in transformers for interpretable classification of retinal images","volume":"82","author":"Playout","year":"2022","journal-title":"Med Image Anal"},{"key":"ref17","first-page":"1","author":"Singh","year":"2023"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"2739","DOI":"10.1109\/JBHI.2022.3214086","article-title":"Multi-label retinal disease classification using transformers","volume":"27","author":"Rodriguez","year":"2023","journal-title":"IEEE J Biomed Health Inform"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"3283","DOI":"10.1007\/s00417-023-06052-x","article-title":"Artificial intelligence in retinal disease: clinical application, challenges, and future directions","volume":"261","author":"Daich Varela","year":"2023","journal-title":"Graefes Arch Clin Exp Ophthalmol"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"150","DOI":"10.3390\/s24010150","article-title":"Multi - scale - denoising residual convolutional network for retinal disease classification using OCT","volume":"24","author":"Peng","year":"2023","journal-title":"Sensors"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"3637","DOI":"10.1038\/s41598-023-30853-z","article-title":"An interpretable transformer network for the retinal disease classification using optical coherence tomography","volume":"13","author":"He","year":"2023","journal-title":"Sci Rep"},{"key":"ref22","first-page":"1","author":"Vishal","year":"2024"},{"key":"ref23","first-page":"934","author":"Nath","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"3459","DOI":"10.1007\/s11517-024-03144-6","article-title":"Multi - label classification of retinal diseases based on fundus images using Resnet and transformer","volume":"62","author":"Zhao","year":"2024","journal-title":"Med Biol Eng Comput"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"5227","DOI":"10.3390\/s24165227","article-title":"The framework of quantifying biomarkers of OCT and OCTA images in retinal diseases","volume":"24","author":"Liu","year":"2024","journal-title":"Sensors (Basel)"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"108726","DOI":"10.1016\/j.compbiomed.2024.108726","article-title":"HTC - retina: A hybrid retinal diseases classification model using transformer - convolutional neural network from optical coherence tomography images","volume":"178","author":"Laouarem","year":"2024","journal-title":"Comput Biol Med"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"140","DOI":"10.1007\/s10103-024-04089-w","article-title":"Automated classification of choroidal neovascularization, diabetic macular edema, and drusen from retinal OCT images using vision transformers: a comparative study","volume":"39","author":"Ak\u00e7a","year":"2024","journal-title":"Lasers Med Sci"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"025033","DOI":"10.1088\/2057-1976\/ad27ea","article-title":"RD - OCT net: hybrid learning system for automated diagnosis of macular diseases from OCT retinal images","volume":"10","author":"Prabha","year":"2024","journal-title":"Biomed Phys Eng Express"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1172","DOI":"10.1364\/boe.6.001172","article-title":"Kernel regression based segmentation of optical coherence tomography images with diabetic macular edema","volume":"6","author":"Chiu","year":"2015","journal-title":"Biomed Opt Express"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"3898750","DOI":"10.1155\/2016\/3898750","article-title":"Multivendor spectral-domain optical coherence tomography dataset, observer annotation performance evaluation, and standardized evaluation framework for intraretinal cystoid fluid segmentation","volume":"2016","author":"Wu","year":"2016","journal-title":"J Ophthalmol"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"989","DOI":"10.1109\/TBME.2017.2734058","article-title":"Fully automated segmentation of fluid\/cyst regions in optical coherence tomography images with diabetic macular edema using neutrosophic sets and graph algorithms","volume":"65","author":"Rashno","year":"2017","journal-title":"IEEE Trans Biomed Eng"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"1122","DOI":"10.1016\/j.cell.2018.02.010","article-title":"Identifying medical diagnoses and treatable diseases by image-based deep learning","volume":"172","author":"Kermany","year":"2018","journal-title":"Cell"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"106532","DOI":"10.1016\/j.compeleceng.2019.106532","article-title":"OCTID: optical coherence tomography image database","volume":"81","author":"Gholami","year":"2020","journal-title":"Comput Electr Eng"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1016\/j.ophtha.2017.10.031","article-title":"Fully automated detection and quantification of macular fluid in OCT using deep learning","volume":"125","author":"Schlegl","year":"2018","journal-title":"Ophthalmology"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"1858","DOI":"10.1109\/TMI.2019.2901398","article-title":"RETOUCH: the retinal OCT fluid detection and segmentation benchmark and challenge","volume":"38","author":"Bogunovi\u0107","year":"2019","journal-title":"IEEE Trans Med Imaging"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"601","DOI":"10.1016\/j.dib.2018.12.073","article-title":"Retinal layer parcellation of optical coherence tomography images: data resource for multiple sclerosis and healthy controls","volume":"22","author":"He","year":"2018","journal-title":"Data Brief"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1016\/j.cmpb.2019.04.027","article-title":"Double-branched and area-constraint fully convolutional networks for automated serous retinal detachment segmentation in SD-OCT images","volume":"176","author":"Gao","year":"2019","journal-title":"Comput Methods Prog Biomed"},{"key":"ref38","first-page":"978","author":"Rao","year":"2019"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1016\/j.media.2019.05.002","article-title":"Automated segmentation of macular edema in OCT using deep neural networks","volume":"55","author":"Hu","year":"2019","journal-title":"Med Image Anal"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"1751","DOI":"10.1364\/osac.387102","article-title":"RMPPNet: residual multiple pyramid pooling network for subretinal fluid segmentation in SD-OCT images","volume":"3","author":"Yang","year":"2020","journal-title":"OSA Contin"},{"key":"ref41","first-page":"793","volume-title":"I. I\u0161gum, B. A. Landman. Medical imaging 2020: Image processing","author":"Bao","year":"2020"},{"key":"ref42","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1038\/s41597-024-03182-7","article-title":"Octdl: optical coherence tomography dataset for image-based deep learning methods","volume":"11","author":"Kulyabin","year":"2024","journal-title":"Sci Data"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1775911\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T05:25:58Z","timestamp":1772688358000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1775911\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,5]]},"references-count":42,"alternative-id":["10.3389\/fmed.2026.1775911"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1775911","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,5]]},"article-number":"1775911"}}