{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T03:02:38Z","timestamp":1787022158534,"version":"build-2736575974"},"reference-count":21,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T00:00:00Z","timestamp":1756771200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100020027","name":"Dr. Rolf M. Schwiete Stiftung","doi-asserted-by":"publisher","award":["2020-024"],"award-info":[{"award-number":["2020-024"]}],"id":[{"id":"10.13039\/501100020027","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    <jats:bold>Purpose<\/jats:bold>\n                    Prediction of the ectasia screening index, an estimator provided by the Casia2 instrument for identifying keratoconus, from raw optical coherence tomography data using convolutional neural networks.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Methods<\/jats:bold>\n                    Three convolutional neural networks models (ResNet18, DenseNet121 and EfficientNetB0) were employed to predict the ectasia screening index. Mean absolute error was used as the performance metric for predicting the ectasia screening index by the adapted convolutional neural network models on the test set. Scans with an ectasia screening index value higher than a certain threshold were classified as\n                    <jats:italic>Keratoconus<\/jats:italic>\n                    , while the remaining scans were classified as\n                    <jats:italic>Not Keratoconus<\/jats:italic>\n                    . The architectures\u2019 performance was evaluated using metrics such as accuracy, sensitivity, specificity, positive predictive value and F1 score on data collected from patients examined at the eye clinic of the Homburg University Hospital. The raw data from the Casia2 instrument, in 3dv format, was converted into 16 images per examination of one eye. For the training, validation and testing phases, 3689, 1050 and 1078 scans (3dv files) were selected, respectively.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Results<\/jats:bold>\n                    In the prediction of the ectasia screening index, the mean absolute error values for the adapted ResNet18, the adapted DenseNet121 and the adapted EfficientNetB0, rounded to two decimal places, were 7.15, 6.64 and 5.86, respectively. In the classification task, the three networks yielded an accuracy of 94.80%, 95.27% and 95.83%, respectively; a sensitivity of 92.07%, 94.64% and 94.17%, respectively; a specificity of 96.61%, 95.69% and 96.92%, respectively; a positive predictive value of 94.72%, 93.55% and 95.28%, respectively; and a F1 score of 93.38%, 94.09% and 94.72%, respectively.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Conclusions<\/jats:bold>\n                    Our results show that the prediction of keratoconus based on the ectasia screening index values estimated from raw data outperforms previous approaches using processed data. adapted EfficientNetB0 outperformed both the other adapted models and those in state-of-the-art studies, with the highest accuracy and F1 score.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0311036","type":"journal-article","created":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T18:09:03Z","timestamp":1756836543000},"page":"e0311036","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":3,"title":["Prediction of the ectasia screening index from raw Casia2 volume data for keratoconus identification by using convolutional neural networks"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-0328-0411","authenticated-orcid":true,"given":"Maziar","family":"Mirsalehi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin","family":"Fassbind","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andreas","family":"Streich","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Achim","family":"Langenbucher","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,9,2]]},"reference":[{"issue":"3","key":"pone.0311036.ref001","doi-asserted-by":"crossref","first-page":"101559","DOI":"10.1016\/j.clae.2021.101559","article-title":"Keratoconus: An updated review","volume":"45","author":"J Santodomingo-Rubido","year":"2022","journal-title":"Cont Lens Anterior Eye"},{"issue":"4","key":"pone.0311036.ref002","doi-asserted-by":"crossref","first-page":"297","DOI":"10.1016\/S0039-6257(97)00119-7","article-title":"Keratoconus","volume":"42","author":"YS Rabinowitz","year":"1998","journal-title":"Surv Ophthalmol"},{"issue":"2","key":"pone.0311036.ref003","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1111\/ceo.13136","article-title":"Applications of corneal topography and tomography: A review","volume":"46","author":"R Fan","year":"2018","journal-title":"Clin Exp Ophthalmol"},{"key":"pone.0311036.ref004","doi-asserted-by":"crossref","unstructured":"Almodin E, Nassaralla BA, Sandes J. Keratoconus: A comprehensive guide to diagnosis and treatment. Springer Nature; 2022.","DOI":"10.1007\/978-3-030-85361-7"},{"key":"pone.0311036.ref005","doi-asserted-by":"crossref","first-page":"108286","DOI":"10.1016\/j.compbiomed.2024.108286","article-title":"CorNet: Autonomous feature learning in raw Corvis ST data for keratoconus diagnosis via residual CNN approach","volume":"172","author":"P Zhang","year":"2024","journal-title":"Comput Biol Med"},{"issue":"10","key":"pone.0311036.ref006","doi-asserted-by":"crossref","first-page":"3898","DOI":"10.1109\/JBHI.2021.3079430","article-title":"KerNet: A novel deep learning approach for keratoconus and sub-clinical keratoconus detection based on raw data of the pentacam HR system","volume":"25","author":"R Feng","year":"2021","journal-title":"IEEE J Biomed Health Inform"},{"issue":"2","key":"pone.0311036.ref007","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0264219","article-title":"Protocol for the diagnosis of keratoconus using convolutional neural networks","volume":"17","author":"J Schatteburg","year":"2022","journal-title":"PLoS One"},{"issue":"1","key":"pone.0311036.ref008","doi-asserted-by":"crossref","first-page":"6566","DOI":"10.1038\/s41598-023-33793-w","article-title":"Automated cornea diagnosis using deep convolutional neural networks based on cornea topography maps","volume":"13","author":"B Fassbind","year":"2023","journal-title":"Sci Rep"},{"key":"pone.0311036.ref009","doi-asserted-by":"crossref","unstructured":"Liu Z, Mao H, Wu CY, Feichtenhofer C, Darrell T, Xie S. A ConvNet for the 2020 s. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition; 2022. p. 11976\u201386.","DOI":"10.1109\/CVPR52688.2022.01167"},{"issue":"12","key":"pone.0311036.ref010","doi-asserted-by":"crossref","first-page":"1046","DOI":"10.1097\/MEG.0b013e3282f198a0","article-title":"Introduction to artificial neural networks","volume":"19","author":"E Grossi","year":"2007","journal-title":"Eur J Gastroenterol Hepatol"},{"issue":"12","key":"pone.0311036.ref011","doi-asserted-by":"crossref","first-page":"6999","DOI":"10.1109\/TNNLS.2021.3084827","article-title":"A survey of convolutional neural networks: Analysis, applications, and prospects","volume":"33","author":"Z Li","year":"2022","journal-title":"IEEE Trans Neural Netw Learning Syst"},{"issue":"3","key":"pone.0311036.ref012","doi-asserted-by":"crossref","first-page":"52","DOI":"10.3390\/computation11030052","article-title":"Theoretical understanding of convolutional neural network: Concepts, architectures, applications, future directions","volume":"11","author":"MM Taye","year":"2023","journal-title":"Computation"},{"key":"pone.0311036.ref013","doi-asserted-by":"crossref","first-page":"1485","DOI":"10.1109\/LSP.2020.3016837","article-title":"On mean absolute error for deep neural network based vector-to-vector regression","volume":"27","author":"J Qi","year":"2020","journal-title":"IEEE Signal Process Lett"},{"issue":"1","key":"pone.0311036.ref014","doi-asserted-by":"crossref","first-page":"5979","DOI":"10.1038\/s41598-022-09954-8","article-title":"On evaluation metrics for medical applications of artificial intelligence","volume":"12","author":"SA Hicks","year":"2022","journal-title":"Sci Rep"},{"key":"pone.0311036.ref015","unstructured":"Python W. Python releases for windows; 2021. p. 24."},{"key":"pone.0311036.ref016","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: 2016 IEEE conference on computer vision and pattern recognition (CVPR); 2016. p. 770\u20138. https:\/\/doi.org\/10.1109\/cvpr.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"pone.0311036.ref017","doi-asserted-by":"crossref","unstructured":"Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2017. p. 4700\u20138.","DOI":"10.1109\/CVPR.2017.243"},{"key":"pone.0311036.ref018","unstructured":"Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. In: International conference on machine learning; 2019. p. 6105\u201314."},{"key":"pone.0311036.ref019","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems; 2019. p. 32."},{"key":"pone.0311036.ref020","first-page":"022022","article-title":"An overview of overfitting and its solutions","volume":"1168","author":"X Ying","year":"2019","journal-title":"J Phys: Conf Ser"},{"issue":"9","key":"pone.0311036.ref021","doi-asserted-by":"crossref","first-page":"6486","DOI":"10.1109\/TPAMI.2024.3382294","article-title":"Towards understanding convergence and generalization of AdamW","volume":"46","author":"P Zhou","year":"2024","journal-title":"IEEE Trans Pattern Anal Mach Intell"}],"updated-by":[{"DOI":"10.1371\/journal.pone.0338609","type":"correction","label":"Correction","source":"publisher","updated":{"date-parts":[[2025,12,9]],"date-time":"2025-12-09T00:00:00Z","timestamp":1765238400000}},{"DOI":"10.1371\/journal.pone.0351460","type":"correction","label":"Correction","source":"publisher","updated":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T00:00:00Z","timestamp":1780963200000}}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0311036","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T18:09:10Z","timestamp":1756836550000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0311036"}},"subtitle":[],"editor":[{"given":"Georgios","family":"Labiris","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,9,2]]},"references-count":21,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2025,9,2]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0311036","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,2]]}}}