{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T18:16:15Z","timestamp":1791483375961,"version":"4.3.4"},"reference-count":35,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2024,4,22]],"date-time":"2024-04-22T00:00:00Z","timestamp":1713744000000},"content-version":"unspecified","delay-in-days":21,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open Ophth"],"accepted":{"date-parts":[[2024,3,28]]},"published-print":{"date-parts":[[2024,4]]},"abstract":"<jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>Our objective is to develop a novel keratoconus image classification system that leverages multiple pretrained models and a transformer architecture to achieve state-of-the-art performance in detecting keratoconus.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Methods and analysis<\/jats:title>\n                  <jats:p>Three pretrained models were used to extract features from the input images. These models have been trained on large datasets and have demonstrated strong performance in various computer vision tasks.<\/jats:p>\n                  <jats:p>The extracted features from the three pretrained models were fused using a feature fusion technique. This fusion aimed to combine the strengths of each model and capture a more comprehensive representation of the input images. The fused features were then used as input to a vision transformer, a powerful architecture that has shown excellent performance in image classification tasks. The vision transformer learnt to classify the input images as either indicative of keratoconus or not.<\/jats:p>\n                  <jats:p>The proposed method was applied to the Shahroud Cohort Eye collection and keratoconus detection dataset. The performance of the model was evaluated using standard evaluation metrics such as accuracy, precision, recall and F1 score.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>The research results demonstrated that the proposed model achieved higher accuracy compared with using each model individually.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>The findings of this study suggest that the proposed approach can significantly improve the accuracy of image classification models for keratoconus detection. This approach can serve as an effective decision support system alongside physicians, aiding in the diagnosis of keratoconus and potentially reducing the need for invasive procedures such as corneal transplantation in severe cases.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1136\/bmjophth-2023-001589","type":"journal-article","created":{"date-parts":[[2024,4,23]],"date-time":"2024-04-23T06:15:11Z","timestamp":1713852911000},"page":"e001589","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":4,"title":["Keratoconus disease classification with multimodel fusion and vision transformer: a pretrained model approach"],"prefix":"10.1136","volume":"9","author":[{"given":"Shokufeh","family":"Yaraghi","sequence":"first","affiliation":[{"name":"Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran (the Islamic Republic of)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5824-9798","authenticated-orcid":false,"given":"Toktam","family":"Khatibi","sequence":"additional","affiliation":[{"name":"Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran (the Islamic Republic of)"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2024,4,22]]},"reference":[{"key":"2025090110091258000_9.1.e001589.1","doi-asserted-by":"publisher","DOI":"10.1016\/S0039-6257(97)00119-7"},{"key":"2025090110091258000_9.1.e001589.2","doi-asserted-by":"publisher","DOI":"10.1016\/0039-6257(84)90094-8"},{"key":"2025090110091258000_9.1.e001589.3","doi-asserted-by":"publisher","DOI":"10.1097\/00004397-199303320-00023"},{"key":"2025090110091258000_9.1.e001589.4","article-title":"Advances in the diagnosis and treatment of Keratoconus","volume":"13","author":"Atalay","year":"2021","journal-title":"Ther Adv Ophthalmol"},{"key":"2025090110091258000_9.1.e001589.5","doi-asserted-by":"crossref","first-page":"2205","DOI":"10.1016\/j.jcrs.2015.10.041","article-title":"Efficacy of axial and tangential corneal topography maps in detecting Subclinical Keratoconus","volume":"41","author":"Tummanapalli","year":"2015","journal-title":"J Cataract Refract Surg"},{"key":"2025090110091258000_9.1.e001589.6","unstructured":"Gurnani B , Kaur K . Keratometer. StatPearls. Treasure Island (FL): StatPearls Publishing, 2023. Available: https:\/\/www.ncbi.nlm.nih.gov\/books\/NBK580516\/"},{"key":"2025090110091258000_9.1.e001589.7","first-page":"21","article-title":"Corneal Waveform measurements has advantages in Pachymetry","volume":"24","author":"Eisner","year":"2006","journal-title":"Ocular Surgery News"},{"key":"2025090110091258000_9.1.e001589.8","doi-asserted-by":"crossref","DOI":"10.1155\/2014\/694542","article-title":"Vision related quality of life in patients with Keratoconus","volume":"2014","author":"Aydin Kurna","year":"2014","journal-title":"J Ophthalmol"},{"key":"2025090110091258000_9.1.e001589.9","doi-asserted-by":"crossref","first-page":"33","DOI":"10.1016\/j.compbiomed.2019.04.024","article-title":"Computer aided diagnosis for suspect Keratoconus detection","volume":"109","author":"Issarti","year":"2019","journal-title":"Comput Biol Med"},{"key":"2025090110091258000_9.1.e001589.10","doi-asserted-by":"crossref","first-page":"101559","DOI":"10.1016\/j.clae.2021.101559","article-title":"Keratoconus: an updated review","volume":"45","author":"Santodomingo-Rubido","year":"2022","journal-title":"Cont Lens Anterior Eye"},{"key":"2025090110091258000_9.1.e001589.11","doi-asserted-by":"crossref","unstructured":"Chen H , Wang Y , Guo T , et al . Pre-trained image processing transformer. In proceedings of the IEEE\/CVF conference on computer vision and pattern recognition. 2021:12299\u2013310.","DOI":"10.1109\/CVPR46437.2021.01212"},{"key":"2025090110091258000_9.1.e001589.12","first-page":"43","article-title":"A survey of shape feature extraction techniques","volume":"15","author":"Mingqiang","year":"2008","journal-title":"Pattern Recognit DAGM"},{"key":"2025090110091258000_9.1.e001589.13","unstructured":"Dosovitskiy A , Beyer L , Kolesnikov A , et al . An image is worth 16X16 words: transformers for image recognition at scale. 2020;arXiv preprint arXiv:2010.11929."},{"key":"2025090110091258000_9.1.e001589.14","doi-asserted-by":"crossref","first-page":"1627","DOI":"10.1007\/s12559-021-09880-3","article-title":"Deep transfer learning for improved detection of Keratoconus using corneal topographic maps","volume":"14","author":"Al-Timemy","year":"2022","journal-title":"Cogn Comput"},{"key":"2025090110091258000_9.1.e001589.15","doi-asserted-by":"crossref","DOI":"10.1167\/tvst.10.14.16","article-title":"A hybrid deep learning construct for detecting Keratoconus from corneal maps","volume":"10","author":"Al-Timemy","year":"2021","journal-title":"Transl Vis Sci Technol"},{"key":"2025090110091258000_9.1.e001589.16","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1167\/tvst.9.2.53","article-title":"Keratoconus screening based on deep learning approach of corneal topography","volume":"9","author":"Kuo","year":"2020","journal-title":"Trans Vis Sci Tech"},{"key":"2025090110091258000_9.1.e001589.17","doi-asserted-by":"crossref","DOI":"10.1155\/2019\/8162567","article-title":"Keratodetect: Keratoconus detection algorithm using Convolutional neural networks","volume":"2019","author":"Lavric","year":"2019","journal-title":"Comput Intell Neurosci"},{"key":"2025090110091258000_9.1.e001589.18","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2019-031313","article-title":"Keratoconus detection using deep learning of colour-coded maps with anterior segment optical coherence tomography: a diagnostic accuracy study","volume":"9","author":"Kamiya","year":"2019","journal-title":"BMJ Open"},{"key":"2025090110091258000_9.1.e001589.19","doi-asserted-by":"crossref","DOI":"10.3390\/diagnostics13101689","article-title":"A deep feature fusion of improved suspected Keratoconus detection with deep learning","volume":"13","author":"Al-Timemy","year":"2023","journal-title":"Diagnostics (Basel)"},{"key":"2025090110091258000_9.1.e001589.20","doi-asserted-by":"crossref","first-page":"519","DOI":"10.1001\/jamaophthalmol.2020.0507","article-title":"Screening candidates for refractive surgery with corneal Tomographic-based deep learning","volume":"138","author":"Xie","year":"2020","journal-title":"JAMA Ophthalmol"},{"key":"2025090110091258000_9.1.e001589.21","doi-asserted-by":"crossref","first-page":"43","DOI":"10.3991\/ijoe.v18i15.33689","article-title":"Transfer learning in Keratoconus classification","volume":"18","author":"Aatila","year":"2022","journal-title":"Int J Onl Eng"},{"key":"2025090110091258000_9.1.e001589.22","doi-asserted-by":"crossref","unstructured":"Otuna-Hern\u00e1ndez DLE-C , Y\u00e1nez-Contreras P , et al . Diagnosis and degree of evolution in a Keratoconus-type corneal Ectasia from image processing. In: Narv\u00e1ez FR , Urgil\u00e9s F , Bastos-Filho TF , et al ., eds. Smart Technologies, Systems and Applications. SmartTech-IC 2022. Communications in Computer and Information Science. 1705. Cham: Springer, 2023. Available: https:\/\/doi.org\/10.1007\/978-3-031-32213-6_21","DOI":"10.1007\/978-3-031-32213-6_21"},{"key":"2025090110091258000_9.1.e001589.23","doi-asserted-by":"crossref","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":"Fassbind","year":"2023","journal-title":"Sci Rep"},{"key":"2025090110091258000_9.1.e001589.24","doi-asserted-by":"crossref","unstructured":"Elsawy A , Abdel-Mottaleb M , Abou Shousha M . Diagnosis of corneal pathologies using deep learning. Ophthalmic Technologies XXX; San Francisco, United States, 2020:150\u201360. doi:10.1117\/12.2552478","DOI":"10.1117\/12.2552478"},{"key":"2025090110091258000_9.1.e001589.25","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1093\/ije\/dyy250","article-title":"Cohort profile: Shahroud schoolchildren eye cohort study (SSCECS)","volume":"48","author":"Emamian","year":"2019","journal-title":"Int J Epidemiol"},{"key":"2025090110091258000_9.1.e001589.26","unstructured":"Perez L , Wang J . The effectiveness of data augmentation in image classification using deep learning. 2017;arXiv preprint arXiv:1712.04621."},{"key":"2025090110091258000_9.1.e001589.27","unstructured":"Baldassarre F , Mor\u00edn DG , Rod\u00e9s-Guirao L . Deep Koalarization: image Colorization using Cnns and inception-Resnet-V2. 2017;arXiv preprint arXiv:1712.03400."},{"key":"2025090110091258000_9.1.e001589.28","unstructured":"Simonyan K , Zisserman A . Very deep Convolutional networks for large-scale image recognition. 2014;arXiv preprint arXiv:1409.1556."},{"key":"2025090110091258000_9.1.e001589.29","doi-asserted-by":"crossref","unstructured":"Koonce B , Koonce B . Efficientnet. In: Convolutional Neural Networks with Swift for Tensorflow: Image Recognition and Dataset Categorization. 2021: 109\u201323.","DOI":"10.1007\/978-1-4842-6168-2_10"},{"key":"2025090110091258000_9.1.e001589.30","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , et al . Deep residual learning for image recognition. In proceedings of the IEEE conference on computer vision and pattern recognition. 2016;770\u20138. doi:10.1109\/CVPR.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"2025090110091258000_9.1.e001589.31","doi-asserted-by":"crossref","unstructured":"Szegedy C , Vanhoucke V , Ioffe S . Rethinking the inception architecture for computer vision. In proceedings of the IEEE conference on computer vision and pattern recognition. IEEE 2016;2818\u201326. doi:10.1109\/CVPR.2016.308","DOI":"10.1109\/CVPR.2016.308"},{"key":"2025090110091258000_9.1.e001589.32","doi-asserted-by":"crossref","unstructured":"Huang G , Liu Z , Van Der maaten L , et al . Densely connected Convolutional networks. In proceedings of the IEEE conference on computer vision and pattern recognition. IEEE 2017;4700\u20138. doi:10.1109\/CVPR.2017.243","DOI":"10.1109\/CVPR.2017.243"},{"key":"2025090110091258000_9.1.e001589.33","unstructured":"Howard AG , Zhu M , Chen B , et al . Mobilenets: efficient Convolutional neural networks for mobile vision applications. 2017;arXiv preprint arXiv:1704.04861."},{"key":"2025090110091258000_9.1.e001589.34","doi-asserted-by":"crossref","unstructured":"Sudha D , Ramakrishna M . Comparative study of features fusion techniques. In 2017 international conference on recent advances in electronics and communication technology (ICRAECT). 2017;235\u20139. doi:10.1109\/ICRAECT.2017.39","DOI":"10.1109\/ICRAECT.2017.39"},{"key":"2025090110091258000_9.1.e001589.35","unstructured":"Han J , Pei J , Kamber M . Data Mining: Concepts and Techniques. Amsterdam, Netherlands: Elsevier, 2011."}],"container-title":["BMJ Open Ophthalmology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjophth-2023-001589","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,1]],"date-time":"2025-09-01T17:09:22Z","timestamp":1756746562000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjophth.bmj.com\/lookup\/doi\/10.1136\/bmjophth-2023-001589"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4]]},"references-count":35,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,4,22]]},"published-print":{"date-parts":[[2024,4]]}},"alternative-id":["10.1136\/bmjophth-2023-001589"],"URL":"https:\/\/doi.org\/10.1136\/bmjophth-2023-001589","relation":{},"ISSN":["2397-3269"],"issn-type":[{"value":"2397-3269","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,4]]}}}