A Deep Learning System Using Optical Coherence Tomography Angiography to Detect Glaucoma and Anterior Ischemic Optic Neuropathy
Journal of Clinical Medicine · 07.01.2023 · 12 (2) · 507
DOI: 10.3390/jcm12020507 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 07.01.2023. Verfasst von Roxane Bunod und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ophthalmology 3, Quinze-Vingts National Ophthalmology Hospital, 28 Rue de Charenton, 75012 Paris, France. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Glaucoma and non-arteritic anterior ischemic optic neuropathy (NAION) are optic neuropathies that can both lead to irreversible blindness. Several studies have compared optical coherence tomography angiography (OCTA) findings in glaucoma and NAION in the presence of similar functional and structural damages with contradictory results. The goal of this study was to use a deep learning system to differentiate OCTA in glaucoma and NAION. Material and methods. Sixty eyes with glaucoma (including primary open angle glaucoma, angle-closure glaucoma, normal tension glaucoma, pigmentary glaucoma, pseudoexfoliative glaucoma and juvenile glaucoma), thirty eyes with atrophic NAION and forty control eyes (NC) were included. All patients underwent OCTA imaging and automatic segmentation was used to analyze the macular superficial capillary plexus (SCP) and the radial peripapillary capillary (RPC) plexus. We used the classic convolutional neural network (CNN) architecture of ResNet50. Attribution maps were obtained using the “Integrated Gradients” method.
Results
The best performances were obtained with the SCP + RPC model achieving a mean area under the receiver operating characteristics curve (ROC AUC) of 0.94 (95% CI 0.92–0.96) for glaucoma, 0.90 (95% CI 0.86–0.94) for NAION and 0.96 (95% CI 0.96–0.97) for NC.
Conclusion
This study shows that deep learning architecture can classify NAION, glaucoma and normal OCTA images with a good diagnostic performance and may outperform the specialist assessment.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Ophthalmology 3, Quinze-Vingts National Ophthalmology Hospital, 28 Rue de Charenton, 75012 Paris, France.
- Keen Eye Technologies SAS, 74 Rue du Faubourg Saint-Antoine, 75012 Paris, France.
- Department of Ophthalmology 4, Quinze-Vingts National Ophthalmology Hospital, 28 Rue de Charenton, 75012 Paris, France.
- IHU FOReSIGHT, INSERM-DGOS CIC 1423, 17 Rue Moreau, 75012 Paris, France.
- INSERM, CNRS, Institut de la Vision, Sorbonne Universités, 7 Rue Moreau, 75012 Paris, France.
- Department of Ophthalmology, Ambroise Paré Hospital, IHU FOReSIGHT, AP-HP, University of Paris Saclay, 9 Avenue Charles de Gaulle, 92100 Boulogne-Billancourt, France.
Zitieren
APA
Bunod, R., Lubrano, M., Pirovano, A., Chotard, G., Brasnu, E., Berlemont, S., Labbé, A., Augstburger, E., & Baudouin, C. (2023). A Deep Learning System Using Optical Coherence Tomography Angiography to Detect Glaucoma and Anterior Ischemic Optic Neuropathy. Journal of Clinical Medicine, 12(2), 507. https://doi.org/10.3390/jcm12020507
Vancouver
Bunod R, Lubrano M, Pirovano A, Chotard G, Brasnu E, Berlemont S, et al. A Deep Learning System Using Optical Coherence Tomography Angiography to Detect Glaucoma and Anterior Ischemic Optic Neuropathy. Journal of Clinical Medicine. 2023;12(2):507. doi:10.3390/jcm12020507
BibTeX
@article{Bunod2023,
author = {Bunod, Roxane and Lubrano, Mélanie and Pirovano, Antoine and Chotard, Géraldine and Brasnu, Emmanuelle and Berlemont, Sylvain and Labbé, Antoine and Augstburger, Edouard and Baudouin, Christophe},
title = {A Deep Learning System Using Optical Coherence Tomography Angiography to Detect Glaucoma and Anterior Ischemic Optic Neuropathy},
journal = {Journal of Clinical Medicine},
year = {2023},
volume = {12},
number = {2},
pages = {507},
doi = {10.3390/jcm12020507}
}
EUVIMED