{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T16:29:57Z","timestamp":1787761797415,"version":"build-2784847793"},"reference-count":42,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"3","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2023,7]]},"abstract":"<jats:sec>\n                    <jats:title>PURPOSE:<\/jats:title>\n                    <jats:p>Automated machine learning (AutoML) allows clinicians without coding experience to build their own deep learning (DL) models. This study assesses the performance of AutoML in diagnosing trachoma from field-collected conjunctival images and compares it to expert-designed DL models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>METHODS:<\/jats:title>\n                    <jats:p>Two ophthalmology trainees without coding experience carried out AutoML model design using a publicly available image data set of field-collected conjunctival images (1656 labeled images). We designed two binary models to differentiate trachomatous inflammation-follicular (TF) and trachomatous inflammation-intense (TI) from normal. We then integrated an Edge model into an Android application using Google Firebase to make offline diagnoses.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>RESULTS:<\/jats:title>\n                    <jats:p>The AutoML models showed high diagnostic properties in the classification tasks that were comparable or better than the bespoke DL models. The TF model had an area under the precision-recall curve (AuPRC) of 0.945, sensitivity of 87%, specificity of 88%, and accuracy of 88%. The TI model had an AuPRC of 0.975, sensitivity of 95%, specificity of 92%, and accuracy of 93%. Through the Android app and using an external dataset, the AutoML model had an AuPRC of 0.875, sensitivity of 83%, specificity of 81%, and accuracy of 83%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>CONCLUSION:<\/jats:title>\n                    <jats:p>AutoML models created by ophthalmologists without coding experience were comparable or better than bespoke models trained on the same dataset. Using AutoML to create models and edge computing to deploy them into smartphone-based apps, our approach brings the whole spectrum of DL model design into the hands of clinicians. This approach has the potential to democratize access to artificial intelligence.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.4103\/sjopt.sjopt_106_22","type":"journal-article","created":{"date-parts":[[2023,5,9]],"date-time":"2023-05-09T01:55:10Z","timestamp":1683597310000},"page":"200-206","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":9,"title":["Development and deployment of a smartphone application for diagnosing trachoma: Leveraging code-free deep learning and edge artificial intelligence"],"prefix":"10.4103","volume":"37","author":[{"given":"Daniel","family":"Milad","sequence":"first","affiliation":[{"name":"Department of Ophthalmology, University of Montreal, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, Maisonneuve-Rosemont Hospital, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, University of Montreal Hospital, Montreal, Quebec, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fares","family":"Antaki","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, University of Montreal, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, Maisonneuve-Rosemont Hospital, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, University of Montreal Hospital, Montreal, Quebec, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marie-Claude","family":"Robert","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, University of Montreal, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, University of Montreal Hospital, Montreal, Quebec, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renaud","family":"Duval","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, University of Montreal, Montreal, Quebec, Canada"},{"name":"Department of Ophthalmology, Maisonneuve-Rosemont Hospital, Montreal, Quebec, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2023,2,16]]},"reference":[{"key":"R2-20240830","doi-asserted-by":"crossref","first-page":"e0009727","DOI":"10.1371\/journal.pntd.0009727","article-title":"Implementing trachoma control programmes in marginalised populations in Tanzania: A qualitative study exploring the experiences and perspectives of key stakeholders","volume":"15","author":"Tsang","year":"2021","journal-title":"PLoS Negl Trop Dis"},{"key":"R3-20240830","doi-asserted-by":"crossref","first-page":"115","DOI":"10.4103\/0974-9233.80699","article-title":"Challenges of trachoma control: An assessment of the situation in Northern Nigeria","volume":"18","author":"Rabiu","year":"2011","journal-title":"Middle East Afr J Ophthalmol"},{"key":"R4-20240830","doi-asserted-by":"crossref","first-page":"e0006023","DOI":"10.1371\/journal.pntd.0006023","article-title":"The cost of mapping trachoma: Data from the global trachoma mapping project","volume":"11","author":"Trotignon","year":"2017","journal-title":"PLoS Negl Trop Dis"},{"key":"R5-20240830","doi-asserted-by":"crossref","first-page":"162","DOI":"10.3109\/09286586.2015.1035792","article-title":"Inter-rater agreement between trachoma graders: Comparison of grades given in field conditions versus grades from photographic review","volume":"22","author":"Gebresillasie","year":"2015","journal-title":"Ophthalmic Epidemiol"},{"key":"R6-20240830","doi-asserted-by":"crossref","first-page":"e0210463","DOI":"10.1371\/journal.pone.0210463","article-title":"Sensitivity and specificity of computer vision classification of eyelid photographs for programmatic trachoma assessment","volume":"14","author":"Kim","year":"2019","journal-title":"PLoS One"},{"key":"R7-20240830","doi-asserted-by":"crossref","first-page":"288","DOI":"10.1038\/s42256-021-00305-2","article-title":"Code-free deep learning for multi-modality medical image classification","volume":"3","author":"Korot","year":"2021","journal-title":"Nat Mach Intell"},{"key":"R8-20240830","doi-asserted-by":"crossref","first-page":"406","DOI":"10.1097\/ICU.0000000000000779","article-title":"Automated deep learning in ophthalmology: AI that can build AI","volume":"32","author":"O\u2019Byrne","year":"2021","journal-title":"Curr Opin Ophthalmol"},{"key":"R9-20240830","doi-asserted-by":"crossref","first-page":"e232","DOI":"10.1016\/S2589-7500(19)30108-6","article-title":"Automated deep learning design for medical image classification by health-care professionals with no coding experience: A feasibility study","volume":"1","author":"Faes","year":"2019","journal-title":"Lancet Digit Health"},{"key":"R10-20240830","doi-asserted-by":"crossref","first-page":"856","DOI":"10.1136\/bjophthalmol-2020-316108","article-title":"Classification of pachychoroid disease on ultrawide-field indocyanine green angiography using auto-machine learning platform","volume":"105","author":"Kim","year":"2021","journal-title":"Br J Ophthalmol"},{"key":"R11-20240830","doi-asserted-by":"crossref","first-page":"2398","DOI":"10.1038\/s41598-022-06127-5","article-title":"Development of a code-free machine learning model for the classification of cataract surgery phases","volume":"12","author":"Touma","year":"2022","journal-title":"Sci Rep"},{"key":"R12-20240830","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1097\/APO.0000000000000398","article-title":"Revisiting the problem of optic nerve detection in a retinal image using automated machine learning","volume":"10","author":"Antaki","year":"2021","journal-title":"Asia Pac J Ophthalmol"},{"key":"R13-20240830","article-title":"Accuracy of automated machine learning in classifying retinal pathologies from ultra-widefield pseudocolour fundus images","author":"Antaki","year":"2021","journal-title":"Br J Ophthalmol"},{"key":"R14-20240830","doi-asserted-by":"crossref","first-page":"19528","DOI":"10.1038\/s41598-020-76665-3","article-title":"Predictive modeling of proliferative vitreoretinopathy using automated machine learning by ophthalmologists without coding experience","volume":"10","author":"Antaki","year":"2020","journal-title":"Sci Rep"},{"key":"R15-20240830","doi-asserted-by":"crossref","first-page":"e000992","DOI":"10.1136\/bmjophth-2022-000992","article-title":"Diagnostic accuracy of code-free deep learning for detection and evaluation of posterior capsule opacification","volume":"7","author":"Huemer","year":"2022","journal-title":"BMJ Open Ophthalmol"},{"key":"R17-20240830","doi-asserted-by":"crossref","first-page":"1111","DOI":"10.1016\/S0140-6736(09)60323-8","article-title":"Assessment of herd protection against trachoma due to repeated mass antibiotic distributions: A cluster-randomised trial","volume":"373","author":"House","year":"2009","journal-title":"Lancet"},{"key":"R18-20240830","first-page":"743","article-title":"A cluster-randomized trial to assess the efficacy of targeting trachoma treatment to children","volume":"64","author":"Amza","year":"2017","journal-title":"Clin Infect Dis"},{"key":"R19-20240830","doi-asserted-by":"crossref","first-page":"e861","DOI":"10.1371\/journal.pntd.0000861","article-title":"Trachoma prevalence and associated risk factors in the gambia and Tanzania: Baseline results of a cluster randomised controlled trial","volume":"4","author":"Harding-Esch","year":"2010","journal-title":"PLoS Negl Trop Dis"},{"key":"R20-20240830","doi-asserted-by":"crossref","first-page":"389","DOI":"10.1186\/s12936-019-3033-2","article-title":"Biannual versus annual mass azithromycin distribution and malaria seroepidemiology among preschool children in Niger: A sub-study of a cluster randomized trial","volume":"18","author":"Oldenburg","year":"2019","journal-title":"Malar J"},{"key":"R21-20240830","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1001\/jamaophthalmol.2013.2356","article-title":"Can we stop mass drug administration prior to 3 annual rounds in communities with low prevalence of trachoma? PRET Ziada trial results","volume":"131","author":"Yohannan","year":"2013","journal-title":"JAMA ophthalmol"},{"key":"R22-20240830","doi-asserted-by":"crossref","first-page":"e0007442","DOI":"10.1371\/journal.pntd.0007442","article-title":"Linear growth in preschool children treated with mass azithromycin distributions for trachoma: A cluster-randomized trial","volume":"13","author":"Keenan","year":"2019","journal-title":"PLoS Negl Trop Dis"},{"key":"R23-20240830","doi-asserted-by":"crossref","first-page":"e1586","DOI":"10.1371\/journal.pntd.0001586","article-title":"Community risk factors for ocular chlamydia infection in Niger: Pre-treatment results from a cluster-randomized trachoma trial","volume":"6","author":"Amza","year":"2012","journal-title":"PLoS Negl Trop Dis"},{"key":"R24-20240830","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1016\/S0140-6736(11)61515-8","article-title":"Comparison of annual versus twice-yearly mass azithromycin treatment for hyperendemic trachoma in Ethiopia: A cluster-randomised trial","volume":"379","author":"Gebre","year":"2012","journal-title":"Lancet"},{"key":"R25-20240830","doi-asserted-by":"crossref","first-page":"e1002633","DOI":"10.1371\/journal.pmed.1002633","article-title":"Mass azithromycin distribution for hyperendemic trachoma following a cluster-randomized trial: A continuation study of randomly reassigned subclusters (TANA II)","volume":"15","author":"Keenan","year":"2018","journal-title":"PLoS Med"},{"key":"R26-20240830","first-page":"477","article-title":"A simple system for the assessment of trachoma and its complications","volume":"65","author":"Thylefors","year":"1987","journal-title":"Bull World Health Organ"},{"key":"R27-20240830","doi-asserted-by":"crossref","first-page":"698","DOI":"10.2471\/BLT.19.248708","article-title":"Modificaci\u00f3n del sistema de clasificaci\u00f3nsimplificada del tracoma","volume":"98","author":"Solomon","year":"2020","journal-title":"BullWorld Health Organ"},{"key":"R28-20240830","article-title":"Underspecification presents challenges for credibility in modern machine learning","author":"D\u2019Amour","year":"2020","journal-title":"Journal of Machine Learning Research"},{"key":"R30-20240830","article-title":"Mobile Operating System Market Share Worldwide","journal-title":"Statcounter Global Stats"},{"key":"R31-20240830","doi-asserted-by":"crossref","first-page":"e51","DOI":"10.1016\/S2589-7500(20)30240-5","article-title":"A global review of publicly available datasets for ophthalmological imaging: Barriers to access, usability, and generalisability","volume":"3","author":"Khan","year":"2021","journal-title":"Lancet Digit Health"},{"key":"R32-20240830","doi-asserted-by":"crossref","first-page":"307","DOI":"10.3389\/fpubh.2017.00307","article-title":"Sensitivity, specificity, and predictive values: Foundations, pliabilities, and pitfalls in research and practice","volume":"5","author":"Trevethan","year":"2017","journal-title":"Front Public Health"},{"key":"R33-20240830","doi-asserted-by":"crossref","first-page":"1350","DOI":"10.1136\/bjophthalmol-2012-302050","article-title":"Application of smartphone cameras for detecting clinically active trachoma","volume":"96","author":"Bhosai","year":"2012","journal-title":"Br J Ophthalmol"},{"key":"R34-20240830","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1145\/1143844.1143874","article-title":"The relationship between precision-recall and ROC curves","author":"Davis","year":"2006","journal-title":"Proceedings of the 23rd International Conference on Machine Learning"},{"key":"R35-20240830","doi-asserted-by":"crossref","first-page":"20","DOI":"10.3109\/09286586.2010.545500","article-title":"Design and baseline data of a randomized trial to evaluate coverage and frequency of mass treatment with azithromycin: The Partnership for Rapid Elimination of Trachoma (PRET) in Tanzania and The Gambia","volume":"18","author":"Stare","year":"2011","journal-title":"Ophthalmic Epidemiol"},{"key":"R36-20240830","doi-asserted-by":"crossref","first-page":"449","DOI":"10.1007\/s40123-020-00275-5","article-title":"The role of telemedicine to alleviate an increasingly burdened healthcare system: Retinopathy of prematurity","volume":"9","author":"Antaki","year":"2020","journal-title":"Ophthalmol Ther"},{"key":"R37-20240830","doi-asserted-by":"crossref","first-page":"1512","DOI":"10.1038\/s41433-020-0913-1","article-title":"Smartphone based ROP (S-ROP) screening-opportunities and challenges","volume":"34","author":"Patil","year":"2020","journal-title":"Eye (Lond)"},{"key":"R38-20240830","doi-asserted-by":"crossref","first-page":"2488","DOI":"10.4269\/ajtmh.20-0386","article-title":"Comparison of smartphone photography, single-lens reflex photography, and field-grading for trachoma","volume":"103","author":"Nesemann","year":"2020","journal-title":"Am J Trop Med Hyg"},{"key":"R39-20240830","doi-asserted-by":"crossref","first-page":"e0009928","DOI":"10.1371\/journal.pntd.0009928","article-title":"Evaluation of photography using head-mounted display technology (ICAPS) for district Trachoma surveys","volume":"15","author":"Naufal","year":"2021","journal-title":"PLoS Negl Trop Dis"},{"key":"R40-20240830","article-title":"The Global Trachoma Mapping Project","journal-title":"Sightsavers"},{"key":"R41-20240830","doi-asserted-by":"crossref","first-page":"880","DOI":"10.1136\/bjo.80.10.880","article-title":"Cost effectiveness and cost utility of preventing trachomatous visual impairment: Lessons from 30 years of trachoma control in Burma","volume":"80","author":"Evans","year":"1996","journal-title":"Br J Ophthalmol"},{"key":"R42-20240830","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1080\/09286580590932761","article-title":"Cost-effectiveness of trachoma control in seven world regions","volume":"12","author":"Baltussen","year":"2005","journal-title":"Ophthalmic Epidemiol"},{"key":"R43-20240830","doi-asserted-by":"crossref","first-page":"613","DOI":"10.1093\/inthealth\/ihz035","article-title":"Smartphone photography as a possible method of post-validation trachoma surveillance in resource-limited settings","volume":"11","author":"Snyder","year":"2019","journal-title":"Int Health"},{"key":"R44-20240830","first-page":"1887","article-title":"Trachoma control: The SAFE strategy","volume":"11","author":"Tian","year":"2018","journal-title":"Int J Ophthalmol"},{"key":"R45-20240830","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1136\/bjophthalmol-2018-313173","article-title":"Artificial intelligence and deep learning in ophthalmology","volume":"103","author":"Ting","year":"2019","journal-title":"Br J Ophthalmol"}],"container-title":["Saudi Journal of Ophthalmology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.ovid.com\/10.4103\/sjopt.sjopt_106_22","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T16:05:33Z","timestamp":1787760333000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ovid.com\/10.4103\/sjopt.sjopt_106_22"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,2,16]]},"references-count":42,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2023]]}},"URL":"https:\/\/doi.org\/10.4103\/sjopt.sjopt_106_22","relation":{},"ISSN":["1319-4534","2542-6680"],"issn-type":[{"value":"1319-4534","type":"print"},{"value":"2542-6680","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,2,16]]},"assertion":[{"value":"2022-07-03","name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-10-09","name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}