Portal de Periódicos da CAPES

Application of a Deep Learning System to Detect Papilledema on Nonmydriatic Ocular Fundus Photographs in an Emergency Department

2023; Elsevier BV; Volume: 261; Linguagem: Inglês

10.1016/j.ajo.2023.10.025

1879-1891

Valérie Biousse, Raymond P. Najjar, Zhiqun Tang, Mung Yan Lin, David W. Wright, Matthew Keadey, Tien Yin Wong, Bonnie Bruce, Dan Miléa, Nancy J. Newman, Clare L. Fraser, Jonathan A. Micieli, Fiona Costello, Étienne Bénard-Séguin, Hui Yang, Carmen Kar Mun Chan, Carol Y. Cheung, Noel C. Y. Chan, Steffen Hamann, Philippe Gohier, Anaïs Vautier, Marie‐Bénédicte Rougier, Christophe Chiquet, Catherine Vignal‐Clermont, Rabih Hage, R.K. Khanna, Thi Hà Châu Tran, Wolf A. Lagrèze, Jost B. Jonas, Ambika Selvakumar, Masoud Aghsaei Fard, Chiara La Morgia, Michele Carbonelli, Piero Barboni, Valério Carelli, Martina Romagnoli, Giulia Amore, Makoto Nakamura, Fumio Takano, Axel Petzold, Maillette de Buy Wenniger lj, Richard C. Kho, Pedro L. Fonseca, Mukharram M. Bikbov, Dan Miléa, Raymond P. Najjar, Daniel Ting, Zhiqun Tang, Jing Liang Loo, Sharon Tow, Shweta Singhal, Caroline Vasseneix, Tien Yin Wong, Ecosse L. Lamoureux, Ching‐Yu Chen, Tin Aung, Leopold Schmetterer, Nicolae Sanda, Gabriele Thuman, Jeong‐Min Hwang, Kavin Vanikieti, Yanin Suwan, Tanyatuth Padungkiatsagul, Patrick Yu‐Wai‐Man, Neringa Jurkutė, Eun Hee Hong, Valérie Biousse, Nancy J. Newman, Jason H. Peragallo, Michael Datillo, Sachin Kedar, Mung Yan Lin, Ajay Patil, Andre Aung, Matthew Boyko, Wael A. Alsakran, A Zayani, Walid Bouthour, Ana Banc, Rasha Mosley, Fernando Labella, Neil R. Miller, John J. Chen, Luis J. Mejico, Janvier Kilangalanga,

Retinal and Optic Conditions

•The Brain and Optic Nerve Study Artificial Intelligence (BONSAI) deep learning system was able to reliably identify papilledema and normal optic discs on photographs obtained in the Fundus photography vs Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) studies.•This deep learning system previously trained on high-quality mydriatic fundus photographs performed very well on nonmydriatic ocular fundus photographs.•Our deep learning system has excellent potential as a diagnostic aid in emergency departments and nonophthalmology clinics equipped with nonmydriatic fundus cameras. PurposeThe Fundus photography vs Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) studies showed that ED providers poorly recognized funduscopic findings in patients in the ED. We tested a modified version of the Brain and Optic Nerve Study Artificial Intelligence (BONSAI) deep learning system on nonmydriatic fundus photographs from the FOTO-ED studies to determine if the deep learning system could have improved the detection of papilledema had it been available to ED providers as a real-time diagnostic aid.DesignRetrospective secondary analysis of a cohort of patients included in the FOTO-ED studies.MethodsThe testing data set included 1608 photographs obtained from 828 patients in the FOTO-ED studies. Photographs were reclassified according to the optic disc classification system used by the deep learning system ("normal optic discs," "papilledema," and "other optic disc abnormalities"). The system's performance was evaluated by calculating the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity using a 1-vs-rest strategy, with reference to expert neuro-ophthalmologists.ResultsThe BONSAI deep learning system successfully distinguished normal from abnormal optic discs (AUC 0.92 [95% confidence interval {CI} 0.90-0.93]; sensitivity 75.6% [73.7%-77.5%] and specificity 89.6% [86.3%-92.8%]), and papilledema from normal and others (AUC 0.97 [0.95-0.99]; sensitivity 84.0% [75.0%-92.6%] and specificity 98.9% [98.5%-99.4%]). Six patients with missed papilledema in 1 eye were correctly identified by the deep learning system as having papilledema in the other eye.ConclusionsThe BONSAI deep learning system was able to reliably identify papilledema and normal optic discs on nonmydriatic photographs obtained in the FOTO-ED studies. Our deep learning system has excellent potential as a diagnostic aid in EDs and non-ophthalmology clinics equipped with nonmydriatic fundus cameras. NOTE: Publication of this article is sponsored by the American Ophthalmological Society. The Fundus photography vs Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) studies showed that ED providers poorly recognized funduscopic findings in patients in the ED. We tested a modified version of the Brain and Optic Nerve Study Artificial Intelligence (BONSAI) deep learning system on nonmydriatic fundus photographs from the FOTO-ED studies to determine if the deep learning system could have improved the detection of papilledema had it been available to ED providers as a real-time diagnostic aid. Retrospective secondary analysis of a cohort of patients included in the FOTO-ED studies. The testing data set included 1608 photographs obtained from 828 patients in the FOTO-ED studies. Photographs were reclassified according to the optic disc classification system used by the deep learning system ("normal optic discs," "papilledema," and "other optic disc abnormalities"). The system's performance was evaluated by calculating the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity using a 1-vs-rest strategy, with reference to expert neuro-ophthalmologists. The BONSAI deep learning system successfully distinguished normal from abnormal optic discs (AUC 0.92 [95% confidence interval {CI} 0.90-0.93]; sensitivity 75.6% [73.7%-77.5%] and specificity 89.6% [86.3%-92.8%]), and papilledema from normal and others (AUC 0.97 [0.95-0.99]; sensitivity 84.0% [75.0%-92.6%] and specificity 98.9% [98.5%-99.4%]). Six patients with missed papilledema in 1 eye were correctly identified by the deep learning system as having papilledema in the other eye. The BONSAI deep learning system was able to reliably identify papilledema and normal optic discs on nonmydriatic photographs obtained in the FOTO-ED studies. Our deep learning system has excellent potential as a diagnostic aid in EDs and non-ophthalmology clinics equipped with nonmydriatic fundus cameras. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.