Deep Learning-Assisted Detection of Glaucoma Progression in Spectral-Domain OCT

Eduardo B Mariottoni; Shounak Datta; Leonardo S Shigueoka; Alessandro A Jammal; Ivan M Tavares; Ricardo Henao; Lawrence Carin; Felipe A Medeiros

doi:10.1016/j.ogla.2022.11.004

Deep Learning-Assisted Detection of Glaucoma Progression in Spectral-Domain OCT

Ophthalmol Glaucoma. 2023 May-Jun;6(3):228-238. doi: 10.1016/j.ogla.2022.11.004. Epub 2022 Nov 18.

Authors

Eduardo B Mariottoni¹, Shounak Datta², Leonardo S Shigueoka³, Alessandro A Jammal³, Ivan M Tavares⁴, Ricardo Henao², Lawrence Carin², Felipe A Medeiros⁵

Affiliations

¹ Vision, Imaging, and Performance (VIP) Laboratory, Duke Eye Center, Duke University, Durham, North Carolina; Department of Ophthalmology, Federal University of São Paulo, São Paulo, Brazil.
² Department of Electrical and Computer Engineering, Pratt School of Engineering, Duke University, Durham, North Carolina.
³ Vision, Imaging, and Performance (VIP) Laboratory, Duke Eye Center, Duke University, Durham, North Carolina.
⁴ Department of Ophthalmology, Federal University of São Paulo, São Paulo, Brazil.
⁵ Vision, Imaging, and Performance (VIP) Laboratory, Duke Eye Center, Duke University, Durham, North Carolina; Department of Electrical and Computer Engineering, Pratt School of Engineering, Duke University, Durham, North Carolina. Electronic address: felipe.medeiros@duke.edu.

Abstract

Purpose: To develop and validate a deep learning (DL) model for detection of glaucoma progression using spectral-domain (SD)-OCT measurements of retinal nerve fiber layer (RNFL) thickness.

Design: Retrospective cohort study.

Participants: A total of 14 034 SD-OCT scans from 816 eyes from 462 individuals.

Methods: A DL convolutional neural network was trained to assess SD-OCT RNFL thickness measurements of 2 visits (a baseline and a follow-up visit) along with time between visits to predict the probability of glaucoma progression. The ground truth was defined by consensus from subjective grading by glaucoma specialists. Diagnostic performance was summarized by the area under the receiver operator characteristic curve (AUC), sensitivity, and specificity, and was compared with conventional trend-based analyses of change. Interval likelihood ratios were calculated to determine the impact of DL model results in changing the post-test probability of progression.

Main outcome measures: The AUC, sensitivity, and specificity of the DL model.

Results: The DL model had an AUC of 0.938 (95% confidence interval [CI], 0.921-0.955), with sensitivity of 87.3% (95% CI, 83.6%-91.6%) and specificity of 86.4% (95% CI, 79.9%-89.6%). When matched for the same specificity, the DL model significantly outperformed trend-based analyses. Likelihood ratios for the DL model were associated with large changes in the probability of progression in the vast majority of SD-OCT tests.

Conclusions: A DL model was able to assess the probability of glaucomatous structural progression from SD-OCT RNFL thickness measurements. The model agreed well with expert judgments and outperformed conventional trend-based analyses of change, while also providing indication of the likely locations of change.

Financial disclosure(s): Proprietary or commercial disclosure may be found after the references.

Keywords: Artificial intelligence; Deep learning; Glaucoma; OCT; Progression.

Publication types

Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

MeSH terms

Deep Learning*
Glaucoma* / diagnosis
Humans
Optic Disk*
Retinal Ganglion Cells
Retrospective Studies
Tomography, Optical Coherence / methods
Visual Fields

Abstract

Publication types

MeSH terms

Grants and funding