{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T03:01:24Z","timestamp":1787022084089,"version":"build-2736575974"},"reference-count":39,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T00:00:00Z","timestamp":1622764800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"by InterDok -- Interdisciplinary Doctoral Studies Projects at Wroclaw University of Science and Technology"},{"DOI":"10.13039\/501100004281","name":"Narodowe Centrum Nauki","doi-asserted-by":"publisher","award":["2018\/29\/B\/ST7\/02451"],"award-info":[{"award-number":["2018\/29\/B\/ST7\/02451"]}],"id":[{"id":"10.13039\/501100004281","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001061","name":"Rebecca L. Cooper Medical Research Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001061","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Health & Medical Research Council Ideas Grant","award":["APP1186915"],"award-info":[{"award-number":["APP1186915"]}]},{"name":"statutory funds of the Department of Systems and Computer Networks, Wroclaw University of Science and Technology."}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:p>\n                    This study aimed to assess the utility of optic nerve head (\n                    <jats:sc>onh<\/jats:sc>\n                    ) en-face images, captured with scanning laser ophthalmoscopy (\n                    <jats:sc>slo<\/jats:sc>\n                    ) during standard optical coherence tomography (\n                    <jats:sc>oct<\/jats:sc>\n                    ) imaging of the posterior segment, and demonstrate the potential of deep learning (\n                    <jats:sc>dl<\/jats:sc>\n                    ) ensemble method that operates in a low data regime to differentiate glaucoma patients from healthy controls. The two groups of subjects were initially categorized based on a range of clinical tests including measurements of intraocular pressure, visual fields,\n                    <jats:sc>oct<\/jats:sc>\n                    derived retinal nerve fiber layer (\n                    <jats:sc>rnfl<\/jats:sc>\n                    ) thickness and dilated stereoscopic examination of\n                    <jats:sc>onh<\/jats:sc>\n                    . 227\n                    <jats:sc>slo<\/jats:sc>\n                    images of 227 subjects (105 glaucoma patients and 122 controls) were used. A new task-specific convolutional neural network architecture was developed for\n                    <jats:sc>slo<\/jats:sc>\n                    image-based classification. To benchmark the results of the proposed method, a range of classifiers were tested including five machine learning methods to classify glaucoma based on\n                    <jats:sc>rnfl<\/jats:sc>\n                    thickness\u2014a well-known biomarker in glaucoma diagnostics, ensemble classifier based on inception v3 architecture, and classifiers based on features extracted from the image. The study shows that cross-validation\n                    <jats:sc>dl<\/jats:sc>\n                    ensemble based on\n                    <jats:sc>slo<\/jats:sc>\n                    images achieved a good discrimination performance with up to 0.962 of balanced accuracy, outperforming all of the other tested classifiers.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0252339","type":"journal-article","created":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T14:12:19Z","timestamp":1622815939000},"page":"e0252339","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":22,"title":["Glaucoma classification based on scanning laser ophthalmoscopic images using a deep learning ensemble method"],"prefix":"10.1371","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0246-3223","authenticated-orcid":true,"given":"Dominika","family":"Su\u0142ot","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Alonso-Caneiro","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pawe\u0142","family":"Ksieniewicz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8034-2534","authenticated-orcid":true,"given":"Patrycja","family":"Krzyzanowska-Berkowska","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"D. Robert","family":"Iskander","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2021,6,4]]},"reference":[{"issue":"3","key":"pone.0252339.ref001","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1136\/bjo.2005.081224","article-title":"The number of people with glaucoma worldwide in 2010 and 2020","volume":"90","author":"HA Quigley","year":"2006","journal-title":"British J. Ophthalmol"},{"key":"pone.0252339.ref002","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1016\/bs.pbr.2015.03.001","article-title":"Strategies to improve early diagnosis in glaucoma","volume":"221","author":"AJ Tatham","year":"2015","journal-title":"Prog. Brain Res"},{"issue":"4","key":"pone.0252339.ref003","doi-asserted-by":"crossref","first-page":"598","DOI":"10.1016\/j.ajo.2007.12.018","article-title":"Role of optic nerve imaging in glaucoma clinical practice and clinical trials","volume":"145","author":"DS Greenfield","year":"2008","journal-title":"Am. J. Ophthalmol"},{"issue":"6","key":"pone.0252339.ref004","doi-asserted-by":"crossref","first-page":"132","DOI":"10.1007\/s10916-016-0482-9","article-title":"Regional image features model for automatic classification between normal and glaucoma in fundus and scanning laser ophthalmoscopy (slo) images","volume":"40","author":"MS Haleem","year":"2016","journal-title":"J. Med. Syst"},{"issue":"8","key":"pone.0252339.ref005","doi-asserted-by":"crossref","first-page":"1557","DOI":"10.1016\/S0161-6420(98)98047-2","article-title":"Identification of early glaucoma cases with the scanning laser ophthalmoscope","volume":"105","author":"G Wollstein","year":"1998","journal-title":"Ophthalmology"},{"issue":"3","key":"pone.0252339.ref006","doi-asserted-by":"crossref","first-page":"254","DOI":"10.1097\/ICU.0000000000000470","article-title":"Deep learning applications in ophthalmology","volume":"29","author":"E Rahimy","year":"2018","journal-title":"Curr. Opin. Ophthalmol"},{"issue":"1","key":"pone.0252339.ref007","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1111\/ceo.13381","article-title":"Current state and future prospects of artificial intelligence in ophthalmology: a review","volume":"47","author":"DT Hogarty","year":"2019","journal-title":"Clin. Experiment. Ophthalmol"},{"key":"pone.0252339.ref008","doi-asserted-by":"crossref","first-page":"134","DOI":"10.1016\/j.neucom.2019.08.079","article-title":"Deep learning based early stage diabetic retinopathy detection using optical coherence tomography","volume":"369","author":"X Li","year":"2019","journal-title":"Neurocomputing"},{"key":"pone.0252339.ref009","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1016\/j.neucom.2020.04.044","article-title":"Retinal optical coherence tomography image classification with label smoothing generative adversarial network","volume":"405","author":"X He","year":"2020","journal-title":"Neurocomputing"},{"issue":"9","key":"pone.0252339.ref010","doi-asserted-by":"crossref","first-page":"963","DOI":"10.1109\/TBME.2002.802012","article-title":"Comparison of machine learning and traditional classifiers in glaucoma diagnosis","volume":"49","author":"K Chan","year":"2002","journal-title":"IEEE. Trans. Biomed. Eng"},{"issue":"6","key":"pone.0252339.ref011","doi-asserted-by":"crossref","first-page":"396","DOI":"10.1097\/OPX.0b013e3181783ab6","article-title":"Machine learning classifiers in glaucoma","volume":"85","author":"C Bowd","year":"2008","journal-title":"Optom. Vis. Sci"},{"issue":"5","key":"pone.0252339.ref012","doi-asserted-by":"crossref","first-page":"e0177726","DOI":"10.1371\/journal.pone.0177726","article-title":"Development of machine learning models for diagnosis of glaucoma","volume":"12","author":"SJ Kim","year":"2017","journal-title":"PLoS One"},{"issue":"10","key":"pone.0252339.ref013","doi-asserted-by":"crossref","first-page":"e003114","DOI":"10.1136\/bmjopen-2013-003114","article-title":"Cross-sectional study: Does combining optical coherence tomography measurements using the \u2018Random Forest\u2019 decision tree classifier improve the prediction of the presence of perimetric deterioration in glaucoma suspects?","volume":"3","author":"K Sugimoto","year":"2013","journal-title":"BMJ Open"},{"key":"pone.0252339.ref014","doi-asserted-by":"crossref","first-page":"136","DOI":"10.1016\/j.ajo.2018.10.007","article-title":"Using deep learning and transfer learning to accurately diagnose early-onset glaucoma from macular optical coherence tomography images","volume":"198","author":"R Asaoka","year":"2019","journal-title":"Am. J. Ophthalmol"},{"issue":"3","key":"pone.0252339.ref015","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1016\/j.ophtha.2019.09.036","article-title":"Deep learning approaches predict glaucomatous visual field damage from OCT optic nerve head en face images and retinal nerve fiber layer thickness maps","volume":"127","author":"M Christopher","year":"2020","journal-title":"Ophthalmology"},{"issue":"7","key":"pone.0252339.ref016","doi-asserted-by":"crossref","first-page":"647","DOI":"10.1097\/IJG.0000000000000988","article-title":"Deep learning classifier with an ultrawide-field scanning laser ophthalmoscope detects glaucoma visual field severity","volume":"27","author":"H Masumoto","year":"2018","journal-title":"J. Glaucoma"},{"issue":"3","key":"pone.0252339.ref017","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1016\/j.bdr.2015.04.001","article-title":"Efficient machine learning for big data: A review","volume":"2","author":"OY Al-Jarrah","year":"2015","journal-title":"Big Data Res"},{"issue":"2","key":"pone.0252339.ref018","doi-asserted-by":"crossref","first-page":"178","DOI":"10.1097\/WNO.0000000000000827","article-title":"Deep learning and transfer learning for optic disc laterality detection: Implications for machine learning in neuro-ophthalmology","volume":"40","author":"TA Liu","year":"2019","journal-title":"J. Neuroophthalmol"},{"issue":"6","key":"pone.0252339.ref019","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1016\/j.ogla.2019.08.004","article-title":"Machine learning models for diagnosing glaucoma from retinal nerve fiber layer thickness maps","volume":"2","author":"P Wang","year":"2019","journal-title":"Ophthalmol. Glaucoma"},{"issue":"10","key":"pone.0252339.ref020","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"SJ Pan","year":"2009","journal-title":"IEEE Trans Knowl. Data Eng"},{"key":"pone.0252339.ref021","unstructured":"Perez L, Wang J. The effectiveness of data augmentation in image classification using deep learning. arXiv preprint arXiv:1712.04621 [Preprint]. 2017 [cited 2021 May 19]. Available from: https:\/\/arxiv.org\/abs\/1712.04621"},{"key":"pone.0252339.ref022","unstructured":"Hinton GE, Srivastava N, Krizhevsky A, Sutskever I, Salakhutdinov RR. Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 [Preprint]. 2012 [cited 2021 May 19]. Available from: https:\/\/arxiv.org\/abs\/1207.0580"},{"key":"pone.0252339.ref023","doi-asserted-by":"crossref","first-page":"150530","DOI":"10.1109\/ACCESS.2019.2947484","article-title":"A deep learning ensemble approach for diabetic retinopathy detection","volume":"7","author":"S Qummar","year":"2019","journal-title":"IEEE Access"},{"issue":"1","key":"pone.0252339.ref024","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s10462-009-9124-7","article-title":"Ensemble-based classifiers","volume":"33","author":"L Rokach","year":"2010","journal-title":"Artif. Intell. Rev"},{"key":"pone.0252339.ref025","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1146\/annurev-bioeng-071516-044442","article-title":"Deep learning in medical image analysis","volume":"19","author":"D Shen","year":"2017","journal-title":"Annu. Rev. Biomed. Eng"},{"issue":"5","key":"pone.0252339.ref026","doi-asserted-by":"crossref","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":"L Fang","year":"2017","journal-title":"Biomed. Opt. Express"},{"issue":"7","key":"pone.0252339.ref027","doi-asserted-by":"crossref","first-page":"3049","DOI":"10.1364\/BOE.9.003049","article-title":"Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers","volume":"9","author":"J Hamwood","year":"2018","journal-title":"Biomed. Opt. Express"},{"issue":"1","key":"pone.0252339.ref028","doi-asserted-by":"crossref","first-page":"6620","DOI":"10.1038\/s41598-017-07103-0","article-title":"Open source software for automatic detection of cone photoreceptors in adaptive optics ophthalmoscopy using convolutional neural networks","volume":"7","author":"D Cunefare","year":"2017","journal-title":"Sci. Rep"},{"issue":"2","key":"pone.0252339.ref029","doi-asserted-by":"crossref","first-page":"413","DOI":"10.1109\/TMI.2019.2927226","article-title":"A large-scale database and a CNN model for attention-based glaucoma detection","volume":"39","author":"L Li","year":"2019","journal-title":"IEEE Trans Med. Imaging"},{"key":"pone.0252339.ref030","unstructured":"Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 [Preprint]. 2014 [cited 2021 May 19]. Available from: https:\/\/arxiv.org\/abs\/1412.6980"},{"key":"pone.0252339.ref031","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume":"9","author":"X Glorot","year":"2010","journal-title":"Proc. AISTATS"},{"key":"pone.0252339.ref032","unstructured":"Chollet F. Keras. 2015. Available from: https:\/\/keras.io"},{"key":"pone.0252339.ref033","unstructured":"Mart\u00edn Abadi et al. TensorFlow: Large-scale machine learning on heterogeneous systems. 2015. Available from tensorflow.org."},{"key":"pone.0252339.ref034","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res"},{"key":"pone.0252339.ref035","doi-asserted-by":"crossref","unstructured":"Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the inception architecture for computer vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016:2818\u20132826.","DOI":"10.1109\/CVPR.2016.308"},{"key":"pone.0252339.ref036","doi-asserted-by":"crossref","DOI":"10.1016\/j.inffus.2013.04.006","article-title":"A survey of multiple classifier systems as hybrid systems","volume":"16","author":"M Wo\u017aniak","year":"2014","journal-title":"Infor. Fusion"},{"key":"pone.0252339.ref037","doi-asserted-by":"crossref","unstructured":"Brodersen KH, Ong CS, Stephan KE, Buhmann JM. The balanced accuracy and its posterior distribution, In 2010 20th International Conference on Pattern Recognition. IEEE. 2010:3121-3124.","DOI":"10.1109\/ICPR.2010.764"},{"key":"pone.0252339.ref038","first-page":"511","volume-title":"Introduction to machine learning","author":"E Alpaydin","year":"2009"},{"key":"pone.0252339.ref039","first-page":"1","article-title":"Constructing synthetic chorio-retinal patches using generative adversarial networks","author":"J Kugelman","year":"2019","journal-title":"Digital Image Computing: Techniques and Applications"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0252339","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,6,4]],"date-time":"2021-06-04T14:13:12Z","timestamp":1622815992000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0252339"}},"subtitle":[],"editor":[{"given":"Demetrios G.","family":"Vavvas","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2021,6,4]]},"references-count":39,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2021,6,4]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0252339","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,6,4]]}}}