Deep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities
dc.contributor.author | Vasseneix C. | |
dc.contributor.author | Nusinovici S. | |
dc.contributor.author | Xu X. | |
dc.contributor.author | Hwang J.M. | |
dc.contributor.author | Hamann S. | |
dc.contributor.author | Chen J.J. | |
dc.contributor.author | Loo J.L. | |
dc.contributor.author | Milea L. | |
dc.contributor.author | Tan K.B.K. | |
dc.contributor.author | Ting D.S.W. | |
dc.contributor.author | Liu Y. | |
dc.contributor.author | Newman N.J. | |
dc.contributor.author | Biousse V. | |
dc.contributor.author | Wong T.Y. | |
dc.contributor.author | Milea D. | |
dc.contributor.author | Najjar R.P. | |
dc.contributor.author | Gohier P. | |
dc.contributor.author | Miller N. | |
dc.contributor.author | Vanikieti K. | |
dc.contributor.author | La Morgia C. | |
dc.contributor.author | Rougier M.B. | |
dc.contributor.author | Ambika S. | |
dc.contributor.author | Fonseca P. | |
dc.contributor.author | Lagrèze W.A. | |
dc.contributor.author | Sanda N. | |
dc.contributor.author | Chiquet C. | |
dc.contributor.author | Yang H. | |
dc.contributor.author | Chan C.K.M. | |
dc.contributor.author | Cheung C.Y. | |
dc.contributor.author | Chau T.T.H. | |
dc.contributor.author | Jurkute N. | |
dc.contributor.author | Yu-Wai-Man P. | |
dc.contributor.author | Kho R. | |
dc.contributor.author | Jonas J.B. | |
dc.contributor.author | Vignal-Clermont C. | |
dc.contributor.author | Kim D.H. | |
dc.contributor.author | Yang H.K. | |
dc.contributor.author | Aung T. | |
dc.contributor.author | Singhal S. | |
dc.contributor.author | Tow S. | |
dc.contributor.author | Nongpiur M.E. | |
dc.contributor.author | Perera S. | |
dc.contributor.author | Narayanaswamy A. | |
dc.contributor.author | Thirugnanam U.N. | |
dc.contributor.author | Fraser C.L. | |
dc.contributor.author | Mejico L.J. | |
dc.contributor.author | Fard M.A. | |
dc.contributor.other | Mahidol University | |
dc.date.accessioned | 2023-06-12T17:16:17Z | |
dc.date.available | 2023-06-12T17:16:17Z | |
dc.date.issued | 2023-06-01 | |
dc.description.abstract | Background: The examination of the optic nerve head (optic disc) is mandatory in patients with headache, hypertension, or any neurological symptoms, yet it is rarely or poorly performed in general clinics. We recently developed a brain and optic nerve study with artificial intelligence-deep learning system (BONSAI-DLS) capable of accurately detecting optic disc abnormalities including papilledema (swelling due to elevated intracranial pressure) on digital fundus photographs with a comparable classification performance to expert neuro-ophthalmologists, but its performance compared to first-line clinicians remains unknown. Methods: In this international, cross-sectional multicenter study, the DLS, trained on 14,341 fundus photographs, was tested on a retrospectively collected convenience sample of 800 photographs (400 normal optic discs, 201 papilledema and 199 other abnormalities) from 454 patients with a robust ground truth diagnosis provided by the referring expert neuro-ophthalmologists. The areas under the receiver-operating-characteristic curves were calculated for the BONSAI-DLS. Error rates, accuracy, sensitivity, and specificity of the algorithm were compared with those of 30 clinicians with or without ophthalmic training (6 general ophthalmologists, 6 optometrists, 6 neurologists, 6 internists, 6 emergency department [ED] physicians) who graded the same testing set of images. Results: With an error rate of 15.3%, the DLS outperformed all clinicians (average error rates 24.4%, 24.8%, 38.2%, 44.8%, 47.9% for general ophthalmologists, optometrists, neurologists, internists and ED physicians, respectively) in the overall classification of optic disc appearance. The DLS displayed significantly higher accuracies than 100%, 86.7% and 93.3% of clinicians (n = 30) for the classification of papilledema, normal, and other disc abnormalities, respectively. Conclusions: The performance of the BONSAI-DLS to classify optic discs on fundus photographs was superior to that of clinicians with or without ophthalmic training. A trained DLS may offer valuable diagnostic aid to clinicians from various clinical settings for the screening of optic disc abnormalities harboring potentially sight- or life-threatening neurological conditions. | |
dc.identifier.citation | Journal of Neuro-Ophthalmology Vol.43 No.2 (2023) , 159-167 | |
dc.identifier.doi | 10.1097/WNO.0000000000001800 | |
dc.identifier.eissn | 15365166 | |
dc.identifier.issn | 10708022 | |
dc.identifier.pmid | 36719740 | |
dc.identifier.scopus | 2-s2.0-85159737109 | |
dc.identifier.uri | https://repository.li.mahidol.ac.th/handle/20.500.14594/83025 | |
dc.rights.holder | SCOPUS | |
dc.subject | Medicine | |
dc.title | Deep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities | |
dc.type | Article | |
mu.datasource.scopus | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85159737109&origin=inward | |
oaire.citation.endPage | 167 | |
oaire.citation.issue | 2 | |
oaire.citation.startPage | 159 | |
oaire.citation.title | Journal of Neuro-Ophthalmology | |
oaire.citation.volume | 43 | |
oairecerif.author.affiliation | Farabi Eye Hospital | |
oairecerif.author.affiliation | Istituto delle Scienze Neurologiche di Bologna | |
oairecerif.author.affiliation | Seoul National University Bundang Hospital | |
oairecerif.author.affiliation | Duke-NUS Medical School | |
oairecerif.author.affiliation | Universidade de Coimbra, Faculdade de Medicina | |
oairecerif.author.affiliation | Fondation Adolphe de Rothschild | |
oairecerif.author.affiliation | Alma Mater Studiorum Università di Bologna | |
oairecerif.author.affiliation | NUS Yong Loo Lin School of Medicine | |
oairecerif.author.affiliation | Medical Research Foundation, Chennai | |
oairecerif.author.affiliation | Universität Freiburg | |
oairecerif.author.affiliation | University of California, Berkeley | |
oairecerif.author.affiliation | Singapore Eye Research Institute | |
oairecerif.author.affiliation | Universite Catholique de Lille | |
oairecerif.author.affiliation | Sun Yat-Sen University | |
oairecerif.author.affiliation | Centro Hospitalar e Universitário de Coimbra | |
oairecerif.author.affiliation | SUNY Upstate Medical University | |
oairecerif.author.affiliation | Singapore General Hospital | |
oairecerif.author.affiliation | Universität Heidelberg | |
oairecerif.author.affiliation | Moorfields Eye Hospital NHS Foundation Trust | |
oairecerif.author.affiliation | Faculty of Medicine Ramathibodi Hospital, Mahidol University | |
oairecerif.author.affiliation | Hong Kong Eye Hospital | |
oairecerif.author.affiliation | Save Sight Institute | |
oairecerif.author.affiliation | UCL Institute of Ophthalmology | |
oairecerif.author.affiliation | Hôpitaux Universitaires de Genève | |
oairecerif.author.affiliation | CHU Angers | |
oairecerif.author.affiliation | Singapore National Eye Centre | |
oairecerif.author.affiliation | Rigshospitalet | |
oairecerif.author.affiliation | Centre Hospitalier Universitaire de Grenoble | |
oairecerif.author.affiliation | Mayo Clinic | |
oairecerif.author.affiliation | A-Star, Institute of High Performance Computing | |
oairecerif.author.affiliation | Groupe Hospitalier Pellegrin | |
oairecerif.author.affiliation | Chinese University of Hong Kong | |
oairecerif.author.affiliation | Emory University School of Medicine | |
oairecerif.author.affiliation | Johns Hopkins School of Medicine | |
oairecerif.author.affiliation | American Eye Center |