OCT Signal Enhancement with Deep Learning

Georgios Lazaridis (1, 2) , Marco Lorenzi (3, 4) , Jibran Mohamed-Noriega (1, 5) , Soledad Aguilar-Munoa (1) , Katsuyoshi Suzuki (1) , Hiroki Nomoto (1) , Sebastien Ourselin (2) , David Garway-Heath (1) , David Crabb , Catey Bunce , Francesca Amalfitano , Nitin Anand , Augusto Azuara-Blanco , Rupert Bourne , David Broadway , Ian Cunliffe , Jeremy Diamond , Scott Fraser , Tuan Ho , Keith Martin , Andrew Mcnaught , Anil Negi , Ameet Shah , Paul Spry , Edward White , Richard Wormald , Wen Xing , Thierry Zeyen

1 UCL - University College of London [London]
2 King‘s College London
3 UniCA - Université Côte d'Azur
4 EPIONE - E-Patient : Images, données & mOdèles pour la médeciNe numériquE
5 UANL - Universidad Autonoma de Nuevo Leon [Mexique]

Purpose To establish whether deep learning methods are able to improve the signal-to-noise ratio of time-domain (TD) OCT images to approach that of spectral-domain (SD) OCT images. Design Method agreement study and progression detection in a randomized, double-masked, placebo-controlled, multicenter trial for open-angle glaucoma (OAG), the United Kingdom Glaucoma Treatment Study (UKGTS). Participants The training and validation cohort comprised 77 stable OAG participants with TD OCT and SD OCT imaging at up to 11 visits within 3 months. The testing cohort comprised 284 newly diagnosed OAG patients with TD OCT images from a cohort of 516 recruited at 10 United Kingdom centers between 2007 and 2010. Methods An ensemble of generative adversarial networks (GANs) was trained on TD OCT and SD OCT image pairs from the training dataset and applied to TD OCT images from the testing dataset. Time-domain OCT images were converted to synthesized SD OCT images and segmented via Bayesian fusion on the output of the GANs. Main Outcome Measures Bland-Altman analysis assessed agreement between TD OCT and synthesized SD OCT average retinal nerve fiber layer thickness (RNFLT) measurements and the SD OCT RNFLT. Analysis of the distribution of the rates of RNFLT change in TD OCT and synthesized SD OCT in the two treatment arms of the UKGTS was compared. A Cox model for predictors of time-to-incident visual field (VF) progression was computed with the TD OCT and the synthesized SD OCT images. Results The 95% limits of agreement were between TD OCT and SD OCT were 26.64 to –22.95; between synthesized SD OCT and SD OCT were 8.11 to –6.73; and between SD OCT and SD OCT were 4.16 to –4.04. The mean difference in the rate of RNFLT change between UKGTS treatment and placebo arms with TD OCT was 0.24 (P = 0.11) and with synthesized SD OCT was 0.43 (P = 0.0017). The hazard ratio for RNFLT slope in Cox regression modeling for time to incident VF progression was 1.09 (95% confidence interval [CI], 1.02–1.21; P = 0.035) for TD OCT and 1.24 (95% CI, 1.08–1.39; P = 0.011) for synthesized SD OCT. Conclusions Image enhancement significantly improved the agreement of TD OCT RNFLT measurements with SD OCT RNFLT measurements. The difference, and its significance, in rates of RNFLT change in the UKGTS treatment arms was enhanced and RNFLT change became a stronger predictor of VF progression.

Mots clés: Deep learning; Glaucoma; Image analysis; OCT; Visual fields

Domaines: Imagerie médicale

Georgios Lazaridis, Marco Lorenzi, Jibran Mohamed-Noriega, Soledad Aguilar-Munoa, Katsuyoshi Suzuki, et al.. OCT Signal Enhancement with Deep Learning. Ophthalmology Glaucoma, 2021, 4 (3), pp.295-304. ⟨10.1016/j.ogla.2020.10.008⟩. ⟨hal-03374545⟩