A Deep Learning Approach for Accurate Discrimination Between Optic Disc Drusen and Papilledema on Fundus Photographs

Kanchalika Sathianvichitr, Raymond P. Najjar, Tang Zhiqun, J. Alexander Fraser, Christine W.L. Yau, Michael J.A. Girard, Fiona Costello, Mung Y. Lin, Wolf A. Lagrèze, Catherine Vignal-Clermont, Clare L. Fraser, Steffen Hamann, Nancy J. Newman, Valérie Biousse, Dan Milea*, Axel Petzold, Philippe Gohier, Neil R. Miller, Kavin Vanikieti, Leonard MileaValerio Carelli, Piero Barboni, Chiara La Morgia, Marie Bénédicte Rougier, Étienne Bénard-Séguin, Selvakumar Ambika, Pedro L. Fonseca, Elisabeth Arnberg Wibroe, Sebastian Kuchlin, Nicolae Sanda, Christophe Chiquet, Hui Yang, Carmen K.M. Chan, Carol Y. Cheung, Janvier Ngoy Kilangalanga, Makoto Nakamura, Takano Fumio, Neringa Jurkute, Patrick Yu-Wai-Man, Richard Kho, Jost Jonas, Marc Joshua Dinkin, John J. Chen, Riccardo Sadun, Jeong Min Hwang, Dong Hyun Kim, Hee Kyung Yang, Jing Liang Loo, Leopold Schmetterer, Ecosse Lamoureux, Tin Aung, Daniel Ting, Tien Yin Wong, Reuben Chao Ming Foo, Shweta Singhal, Sharon Lee Choon Tow, Caroline Vasseneix, Luis J. Mejico, Masoud Aghsaei Fard, Jonathan A. Micieli, Mukharram M. Bikbov

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Background:Optic disc drusen (ODD) represent an important differential diagnosis of papilledema caused by intracranial hypertension, but their distinction may be difficult in clinical practice. The aim of this study was to train, validate, and test a dedicated deep learning system (DLS) for binary classification of ODD vs papilledema (including various subgroups within each category), on conventional mydriatic digital ocular fundus photographs collected in a large international multiethnic population.Methods:This retrospective study included 4,508 color fundus images in 2,180 patients from 30 neuro-ophthalmology centers (19 countries) participating in the Brain and Optic Nerve Study with Artificial Intelligence (BONSAI) Group. For training and internal validation, we used 857 ODD images and 3,230 papilledema images, in 1,959 patients. External testing was performed on an independent data set (221 patients), including 207 images with ODD (96 visible and 111 buried), provided by 3 centers of the Optic Disc Drusen Studies Consortium, and 214 images of papilledema (92 mild-to-moderate and 122 severe) from a previously validated study.Results:The DLS could accurately distinguish between all ODD and papilledema (all severities included): Area under the receiver operating characteristic curve (AUC) 0.97 (95% confidence interval [CI], 0.96-0.98), accuracy 90.5% (95% CI, 88.0%-92.9%), sensitivity 86.0% (95% CI, 82.1%-90.1%), and specificity 94.9% (95% CI, 92.3%-97.6%). The performance of the DLS remained high for discrimination of buried ODD from mild-to-moderate papilledema: AUC 0.93 (95% CI, 0.90-0.96), accuracy 84.2% (95% CI, 80.2%-88.6%), sensitivity 78.4% (95% CI, 72.2%-84.7%), and specificity 91.3% (95% CI, 87.0%-96.4%).Conclusions:A dedicated DLS can accurately distinguish between ODD and papilledema caused by intracranial hypertension, even when considering buried ODD vs mild-to-moderate papilledema.

Original languageEnglish
Pages (from-to)454-461
Number of pages8
JournalJournal of Neuro-Ophthalmology
Volume44
Issue number4
DOIs
Publication statusPublished - Dec 1 2024

Bibliographical note

Publisher Copyright:
© North American Neuro-Ophthalmology Society.

ASJC Scopus Subject Areas

  • Ophthalmology
  • Clinical Neurology

Fingerprint

Dive into the research topics of 'A Deep Learning Approach for Accurate Discrimination Between Optic Disc Drusen and Papilledema on Fundus Photographs'. Together they form a unique fingerprint.

Cite this