{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,19]],"date-time":"2026-07-19T19:54:08Z","timestamp":1784490848359,"version":"3.55.0"},"reference-count":23,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2022,9,21]],"date-time":"2022-09-21T00:00:00Z","timestamp":1663718400000},"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. Public Health"],"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>To apply deep learning (DL) techniques to develop an automatic intelligent classification system identifying the specific types of myopic maculopathy (MM) based on macular optical coherence tomography (OCT) images using transfer learning (TL).<\/jats:p><\/jats:sec><jats:sec><jats:title>Method<\/jats:title><jats:p>In this retrospective study, a total of 3,945 macular OCT images from 2,866 myopic patients were recruited from the ophthalmic outpatients of three hospitals. After culling out 545 images with poor quality, a dataset containing 3,400 macular OCT images was manually classified according to the ATN system, containing four types of MM with high OCT diagnostic values. Two DL classification algorithms were trained to identify the targeted lesion categories: Algorithm A was trained from scratch, and algorithm B using the TL approach initiated from the classification algorithm developed in our previous study. After comparing the training process, the algorithm with better performance was tested and validated. The performance of the classification algorithm in the test and validation sets was evaluated using metrics including sensitivity, specificity, accuracy, quadratic-weighted kappa score, and the area under the receiver operating characteristic curve (AUC). Moreover, the human-machine comparison was conducted. To better evaluate the algorithm and clarify the optimization direction, the dimensionality reduction analysis and heat map analysis were also used to visually analyze the algorithm.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Algorithm B showed better performance in the training process. In the test set, the algorithm B achieved relatively robust performance with macro AUC, accuracy, and quadratic-weighted kappa of 0.986, 96.04% (95% CI: 0.951, 0.969), and 0.940 (95% CI: 0.909\u20130.971), respectively. In the external validation set, the performance of algorithm B was slightly inferior to that in the test set. In human-machine comparison test, the algorithm indicators were inferior to the retinal specialists but were the same as the ordinary ophthalmologists. In addition, dimensionality reduction visualization and heatmap visualization analysis showed excellent performance of the algorithm.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>Our macular OCT image classification algorithm developed using the TL approach exhibited excellent performance. The automatic diagnosis system for macular OCT images of MM based on DL showed potential application prospects.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fpubh.2022.1005700","type":"journal-article","created":{"date-parts":[[2022,9,21]],"date-time":"2022-09-21T05:20:02Z","timestamp":1663737602000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":16,"title":["Development of a deep learning algorithm for myopic maculopathy classification based on OCT images using transfer learning"],"prefix":"10.3389","volume":"10","author":[{"given":"Xiaoying","family":"He","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peifang","family":"Ren","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li","family":"Lu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuyuan","family":"Tang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zixuan","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2022,9,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"50","DOI":"10.3109\/09286580903450346","article-title":"Prevalence of visual impairment in the adult Japanese population by cause and severity and future projections","volume":"17","author":"Yamada","year":"2010","journal-title":"Ophthalmic Epidemiol."},{"key":"B2","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1097\/ICL.0000000000000223","article-title":"Complications of pathologic myopia","volume":"42","author":"Cho","year":"2016","journal-title":"Eye Contact Lens."},{"key":"B3","article-title":"Long-term pattern of progression of myopic maculopathy: a natural history study","volume-title":"Ophthalmology.","author":"Hayashi","year":"2010"},{"key":"B4","doi-asserted-by":"publisher","first-page":"877","DOI":"10.1016\/j.ajo.2015.01.022","article-title":"International photographic classification and grading system for myopic maculopathy","volume":"159","author":"Ohno-Matsui","year":"2015","journal-title":"Am J Ophthalmol."},{"key":"B5","doi-asserted-by":"publisher","first-page":"33771","DOI":"10.1109\/ACCESS.2019.2891975","article-title":"Integrating handcrafted and deep features for optical coherence tomography based retinal disease classification","volume":"7","author":"Li","year":"2019","journal-title":"IEEE Access."},{"key":"B6","doi-asserted-by":"publisher","first-page":"2732","DOI":"10.1364\/BOE.8.002732","article-title":"Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search","volume":"8","author":"Fang","year":"2017","journal-title":"Biomed Opt Express."},{"key":"B7","doi-asserted-by":"publisher","first-page":"3440","DOI":"10.1364\/BOE.8.003440","article-title":"Deep-learning based, automated segmentation of macular edema in optical coherence tomography","volume":"8","author":"Lee","year":"2017","journal-title":"Biomed Opt Express."},{"key":"B8","doi-asserted-by":"publisher","first-page":"322","DOI":"10.1016\/j.oret.2016.12.009","article-title":"Deep learning is effective for the classification of OCT images of normal versus age-related macular degeneration","volume":"1","author":"Lee","year":"2017","journal-title":"Ophthalmol Retina."},{"key":"B9","doi-asserted-by":"publisher","first-page":"3244","DOI":"10.1364\/BOE.9.003244","article-title":"DRUNET: a dilated-residual U-Net deep learning network to segment optic nerve head tissues in optical coherence tomography images","volume":"9","author":"Devalla","year":"2018","journal-title":"Biomed Opt Express."},{"key":"B10","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":"Comp Meth Prog Biomed."},{"key":"B11","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1136\/bjophthalmol-2020-317825","article-title":"Development and validation of a deep learning system to screen vision-threatening conditions in high myopia using optical coherence tomography images","volume":"106","author":"Li","year":"2020","journal-title":"Clin Sci."},{"key":"B12","doi-asserted-by":"publisher","first-page":"652848","DOI":"10.3389\/fcell.2021.652848","article-title":"An optical coherence tomography-based deep learning algorithm for visual acuity prediction of highly myopic eyes after cataract surgery","volume":"9","author":"Wei","year":"2021","journal-title":"Front Cell Dev Biol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1167\/tvst.10.13.10","article-title":"Automatic screening and identifying myopic maculopathy on optical coherence tomography images using deep learning","volume":"10","author":"Ye","year":"2021","journal-title":"Transl Vis Sci Technol."},{"key":"B14","doi-asserted-by":"publisher","first-page":"80","DOI":"10.1016\/j.preteyeres.2018.10.005","article-title":"Myopic maculopathy: current status and proposal for a new classification and grading system (ATN)","volume":"69","author":"Ruiz-Medrano","year":"2019","journal-title":"Prog Retin Eye Res."},{"key":"B15","doi-asserted-by":"publisher","first-page":"719262","DOI":"10.3389\/fcell.2021.719262","article-title":"AI-Model for identifying pathologic myopia based on deep learning algorithms of myopic maculopathy classification and \u201cPlus\u201d lesion detection in fundus images","volume":"9","author":"Lu","year":"2021","journal-title":"Front Cell Dev Biol."},{"key":"B16","doi-asserted-by":"publisher","first-page":"1225","DOI":"10.1038\/s42003-021-02758-y","article-title":"Development of deep learning-based detecting systems for pathologic myopia using retinal fundus images","volume":"4","author":"Lu","year":"2021","journal-title":"Commun Biol."},{"key":"B17","doi-asserted-by":"publisher","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis","year":"1977","journal-title":"Int Biomet Soc."},{"key":"B18","doi-asserted-by":"publisher","first-page":"1437","DOI":"10.1038\/eye.2014.221","article-title":"Multimodal imaging including spectral-domain optical coherence tomography and confocal near-infrared reflectance for characterization of lacquer cracks in highly myopic eyes","volume":"28","author":"Liu","year":"2014","journal-title":"Eye (Lond)."},{"key":"B19","doi-asserted-by":"publisher","first-page":"586","DOI":"10.1038\/s41598-020-79809-7","article-title":"Development and validation of deep learning algorithms for automated eye laterality detection with anterior segment photography","volume":"11","author":"Zheng","year":"2021","journal-title":"Sci Rep."},{"key":"B20","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Comp Soc."},{"key":"B21","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature."},{"key":"B22","doi-asserted-by":"publisher","first-page":"356","DOI":"10.1016\/j.ajo.2019.07.010","article-title":"Morphological characteristics and risk factors of myopic maculopathy in an older high myopia population-based on the new classification system (ATN)","volume":"208","author":"Chen","year":"2019","journal-title":"Am J Ophthalmol."},{"key":"B23","doi-asserted-by":"publisher","first-page":"907","DOI":"10.1167\/iovs.16-20742","article-title":"Is choroidal or scleral thickness related to myopic macular degeneration?","volume":"58","author":"Wong","year":"2017","journal-title":"Invest Ophthalmol Vis Sc."}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2022.1005700\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,21]],"date-time":"2022-09-21T05:20:04Z","timestamp":1663737604000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2022.1005700\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9,21]]},"references-count":23,"alternative-id":["10.3389\/fpubh.2022.1005700"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2022.1005700","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,9,21]]},"article-number":"1005700"}}