{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T19:29:44Z","timestamp":1774380584973,"version":"3.50.1"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T00:00:00Z","timestamp":1774310400000},"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. Pediatr."],"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Retinopathy of prematurity (ROP) is a leading cause of blindness in infants. Early and accurate screening is essential. Current deep learning systems can help, yet their accuracy drops when used on different population. We aimed to find the best deep learning model for plus disease in ROP and to test it on a multi-center dataset.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We built a cross-center retinal image database by merging public and private sets (FARFUM-RoP, HVDROPDB, LAN-RoP, Preterm infants &amp;lt;34 weeks GA from three tertiary NICUs, 2635 images). Nine types models were compared: ResNet34, ResNet50, DenseNet, Inception, MobileNet, VGG16, VGG19, EfficientNet, and Swin-Transformer. We compared FLOPs, parameter count, accuracy, recall, precision, and F1 score. ResNet50 showed the best balance and was kept. Ten-fold cross-validation was run on FARFUM-RoP alone, LAN-RoP alone, and their combined set.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Across the three diagnostic tasks, the ResNet50 algorithm attained area-under-the-curve (AUC) values of 0.97, 0.95 and 1.00 (95% CI 0.94\u20130.99, 0.91\u20130.98, 0.97\u20131.00) for Normal, Pre-plus and Plus disease, respectively. When trained on the consolidated multi-centre cohort, the model achieved optimal overall performance, delivering an accuracy of 92.60%, recall of 92.58%, precision of 92.69% and F1-score of 92.60%\u2014all metrics surpassing those obtained with any single-centre training set.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Compared with single-centre training, the cross-centre fusion strategy significantly enhanced the generalisability of the artificial-intelligence model, yielded superior diagnostic indices, and improved diagnostic accuracy for infants from diverse demographic backgrounds.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fped.2026.1765353","type":"journal-article","created":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T16:10:38Z","timestamp":1774368638000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Research on the construction of an AI diagnostic model for plus disease of retinopathy of prematurity based on cross-center fusion datasets"],"prefix":"10.3389","volume":"14","author":[{"given":"Xiqianru","family":"Zhang","sequence":"first","affiliation":[{"name":"The First School of Clinical Medicine, Lanzhou University","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huichun","family":"Liang","sequence":"additional","affiliation":[{"name":"School of Medical Informatics and Engineering, Gansu University of Chinese Medicine","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ruifeng","family":"Wang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Dunhuang Medical and Transformation, Ministry of Education of the People\u2019s Republic of China, Gansu University of Chinese Medicine","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rouqing","family":"Wu","sequence":"additional","affiliation":[{"name":"The First School of Clinical Medicine, Lanzhou University","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiao","family":"Shen","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, The First Hospital of Lanzhou University","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuemei","family":"Zhang","sequence":"additional","affiliation":[{"name":"The First School of Clinical Medicine, Lanzhou University","place":["Lanzhou, Gansu, China"]},{"name":"Department of Ophthalmology, The First Hospital of Lanzhou University","place":["Lanzhou, Gansu, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,3,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"791","DOI":"10.1001\/jamaophthalmol.2022.2135","article-title":"External validation of a retinopathy of prematurity screening model using artificial intelligence in 3 low- and middle-income populations","volume":"140","author":"Coyner","year":"2022","journal-title":"JAMA Ophthalmol"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1383","DOI":"10.1001\/jamaophthalmol.2018.4290","article-title":"Incidence and early course of retinopathy of prematurity: secondary analysis of the postnatal growth and retinopathy of prematurity (G-ROP) study","volume":"136","author":"Quinn","year":"2018","journal-title":"JAMA Ophthalmol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"138","DOI":"10.1080\/09286586.2020.1793372","article-title":"What is the incidence of retinopathy of prematurity (ROP). In \u201cbig\u201d babies?: results of a retrospective multicenter study","volume":"28","author":"Ekinci","year":"2021","journal-title":"Ophthalmic Epidemiol"},{"key":"B4","doi-asserted-by":"publisher","first-page":"19401","DOI":"10.1038\/s41598-020-76472-w","article-title":"Survival rate dependent variations in retinopathy of prematurity treatment rates in very low birth weight infants","volume":"10","author":"Park","year":"2020","journal-title":"Sci Rep"},{"key":"B5","doi-asserted-by":"publisher","first-page":"803","DOI":"10.1001\/jamaophthalmol.2018.1934","article-title":"Automated diagnosis of plus disease in retinopathy of prematurity using deep convolutional neural networks","volume":"136","author":"Brown","year":"2018","journal-title":"JAMA Ophthalmol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"e339","DOI":"10.1111\/aos.14264","article-title":"Automated diagnosis and quantitative analysis of plus disease in retinopathy of prematurity based on deep convolutional neural networks","volume":"98","author":"Mao","year":"2020","journal-title":"Acta Ophthalmol (Copenh)"},{"key":"B7","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1007\/s10916-020-01635-1","article-title":"Automatic grading of retinal blood vessel in deep retinal image diagnosis","volume":"44","author":"Maji","year":"2020","journal-title":"J Med Syst"},{"key":"B8","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1167\/tvst.8.6.23","article-title":"Deep learning algorithm for automated diagnosis of retinopathy of prematurity plus disease","volume":"8","author":"Tan","year":"2019","journal-title":"Transl Vis Sci Technol"},{"key":"B9","doi-asserted-by":"publisher","first-page":"932","DOI":"10.1007\/s10278-021-00477-8","article-title":"Assistive framework for automatic detection of all the zones in retinopathy of prematurity using deep learning","volume":"34","author":"Agrawal","year":"2021","journal-title":"J Digit Imaging"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1029","DOI":"10.1001\/jamaophthalmol.2019.2442","article-title":"A quantitative severity scale for retinopathy of prematurity using deep learning to monitor disease regression after treatment","volume":"137","author":"Gupta","year":"2019","journal-title":"JAMA Ophthalmol"},{"key":"B11","doi-asserted-by":"publisher","first-page":"3137","DOI":"10.1007\/s00417-021-05191-3","article-title":"Effect of gestational age at birth, sex, and race on foveal structure in children","volume":"259","author":"Jin","year":"2021","journal-title":"Graefe\u2019s Arch Clin Exp Ophthalmol Albrecht von Graefes Archiv fur Klinische und Experimentelle Ophthalmologie"},{"key":"B12","doi-asserted-by":"publisher","first-page":"859310","DOI":"10.3389\/fcvm.2022.859310","article-title":"Fairness in cardiac magnetic resonance imaging: assessing sex and racial bias in deep learning-based segmentation","volume":"9","author":"Puyol-Ant\u00f3n","year":"2022","journal-title":"Front Cardiovasc Med"},{"key":"B13","doi-asserted-by":"publisher","first-page":"631","DOI":"10.1007\/s10729-024-09691-6","article-title":"Evaluating machine learning model bias and racial disparities in non-small cell lung cancer using SEER registry data","volume":"27","author":"Trentz","year":"2024","journal-title":"Health Care Manag Sci"},{"key":"B14","doi-asserted-by":"publisher","first-page":"e2421290","DOI":"10.1001\/jamanetworkopen.2024.21290","article-title":"Fairness in predicting cancer mortality across racial subgroups","volume":"7","author":"Ganta","year":"2024","journal-title":"JAMA Netw Open"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1172","DOI":"10.1093\/jamia\/ocae060","article-title":"Unmasking bias in artificial intelligence: a systematic review of bias detection and mitigation strategies in electronic health record-based models","volume":"31","author":"Chen","year":"2024","journal-title":"J Am Med Inform Assoc JAMIA"},{"key":"B16","doi-asserted-by":"publisher","first-page":"e60269","DOI":"10.2196\/60269","article-title":"Bias mitigation in primary health care artificial intelligence models: scoping review","volume":"27","author":"Sasseville","year":"2025","journal-title":"J Med Internet Res"},{"key":"B17","doi-asserted-by":"publisher","first-page":"e70023","DOI":"10.1002\/mco2.70023","article-title":"Multimodal data fusion AI model uncovers tumor microenvironment immunotyping heterogeneity and enhanced risk stratification of breast cancer","volume":"5","author":"Yu","year":"2024","journal-title":"MedComm"},{"key":"B18","doi-asserted-by":"publisher","first-page":"104833","DOI":"10.1016\/j.jbi.2025.104833","article-title":"Evaluating an information theoretic approach for selecting multimodal data fusion methods","volume":"167","author":"Zhang","year":"2025","journal-title":"J Biomed Inform"},{"key":"B19","doi-asserted-by":"publisher","first-page":"jjae197","DOI":"10.1093\/ecco-jcc\/jjae197","article-title":"Multi-omics data integration identifies novel biomarkers and patient subgroups in inflammatory bowel disease","volume":"19","author":"Preto","year":"2025","journal-title":"J Crohn\u2019s Colitis"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e2420","DOI":"10.7717\/peerj-cs.2420","article-title":"Locustlens: leveraging environmental data fusion and machine learning for desert locust swarm prediction","volume":"10","author":"Khan","year":"2024","journal-title":"PeerJ Comp Sci"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1176","DOI":"10.1038\/s41597-024-03897-7","article-title":"FARFUM-RoP, a dataset for computer-aided detection of retinopathy of prematurity","volume":"11","author":"Akbari","year":"2024","journal-title":"Sci Data"},{"key":"B22","doi-asserted-by":"publisher","first-page":"e51","DOI":"10.1016\/j.ophtha.2021.05.031","article-title":"International classification of retinopathy of prematurity, third edition","volume":"128","author":"Chiang","year":"2021","journal-title":"Ophthalmology"},{"key":"B23","doi-asserted-by":"publisher","first-page":"109839","DOI":"10.1016\/j.dib.2023.109839","article-title":"HVDROPDB datasets for research in retinopathy of prematurity","volume":"52","author":"Agrawal","year":"2024","journal-title":"Data Brief"},{"key":"B24","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1186\/s12911-024-02518-y","article-title":"A hybrid framework for glaucoma detection through federated machine learning and deep learning models","volume":"24","author":"Aljohani","year":"2024","journal-title":"BMC Med Inform Decis Mak"},{"key":"B25","doi-asserted-by":"publisher","first-page":"e73985","DOI":"10.7759\/cureus.73985","article-title":"Ocular duction measurement using three convolutional neural network models: a comparative study","volume":"16","author":"Chuntranapaporn","year":"2024","journal-title":"Cureus"},{"key":"B26","doi-asserted-by":"publisher","first-page":"e15734056286918","DOI":"10.2174\/0115734056286918240419100058","article-title":"An artificial intelligence driven approach for classification of ophthalmic images using convolutional neural network: an experimental study","volume":"20","author":"Singh","year":"2024","journal-title":"Curr Med Imaging"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1760","DOI":"10.1016\/j.ultrasmedbio.2023.03.022","article-title":"Establishment and evaluation of intelligent diagnostic model for ophthalmic ultrasound images based on deep learning","volume":"49","author":"Li","year":"2023","journal-title":"Ultrasound Med Biol"},{"key":"B28","doi-asserted-by":"publisher","first-page":"100552","DOI":"10.1016\/j.xops.2024.100552","article-title":"Comparative analysis of vision transformers and conventional convolutional neural networks in detecting referable diabetic retinopathy","volume":"4","author":"Goh","year":"2024","journal-title":"Ophthalmol Sci"},{"key":"B29","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1016\/j.jclinepi.2023.07.020","article-title":"The walking man approach to interpreting the receiver operating characteristic curve and area under the receiver operating characteristic curve","volume":"162","author":"Kohn","year":"2023","journal-title":"J Clin Epidemiol"},{"key":"B30","doi-asserted-by":"publisher","first-page":"855","DOI":"10.1016\/j.jclinepi.2015.02.010","article-title":"The precision\u2013recall curve overcame the optimism of the receiver operating characteristic curve in rare diseases","volume":"68","author":"Ozenne","year":"2015","journal-title":"J Clin Epidemiol"},{"key":"B31","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD\u2009+\u2009AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"Br Med J"},{"key":"B32","doi-asserted-by":"publisher","first-page":"102554","DOI":"10.1016\/j.mex.2024.102554","article-title":"Performance analysis of hybrid deep learning framework using a vision transformer and convolutional neural network for handwritten digit recognition","volume":"12","author":"Agrawal","year":"2024","journal-title":"MethodsX"},{"key":"B33","doi-asserted-by":"publisher","first-page":"e0233514","DOI":"10.1371\/journal.pone.0233514","article-title":"A convolutional neural network for the screening and staging of diabetic retinopathy","volume":"15","author":"Shaban","year":"2020","journal-title":"PLoS One"},{"key":"B34","doi-asserted-by":"publisher","first-page":"106294","DOI":"10.1016\/j.cmpb.2021.106294","article-title":"A novel multiscale and multipath convolutional neural network based age-related macular degeneration detection using OCT images","volume":"209","author":"Thomas","year":"2021","journal-title":"Comput Methods Programs Biomed"}],"container-title":["Frontiers in Pediatrics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2026.1765353\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T16:10:40Z","timestamp":1774368640000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2026.1765353\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,24]]},"references-count":34,"alternative-id":["10.3389\/fped.2026.1765353"],"URL":"https:\/\/doi.org\/10.3389\/fped.2026.1765353","relation":{},"ISSN":["2296-2360"],"issn-type":[{"value":"2296-2360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,24]]},"article-number":"1765353"}}