Detection of Nonexudative Macular Neovascularization on Structural OCT Images Using Vision Transformers

Yuka Kihara; Mengxi Shen; Yingying Shi; Xiaoshuang Jiang; Liang Wang; Rita Laiginhas; Cancan Lyu; Jin Yang; Jeremy Liu; Rosalyn Morin; Randy Lu; Hironobu Fujiyoshi; William J Feuer; Giovanni Gregori; Philip J Rosenfeld; Aaron Y Lee

doi:10.1016/j.xops.2022.100197

Detection of Nonexudative Macular Neovascularization on Structural OCT Images Using Vision Transformers

Ophthalmol Sci. 2022 Jul 8;2(4):100197. doi: 10.1016/j.xops.2022.100197. eCollection 2022 Dec.

Authors

Affiliations

¹ University of Washington, Department of Ophthalmology, Seattle, Washington.
² Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida.
³ Department of Robotic Science and Technology, Chubu University, Aichi, Japan.

Abstract

Purpose: A deep learning model was developed to detect nonexudative macular neovascularization (neMNV) using OCT B-scans.

Design: Retrospective review of a prospective, observational study.

Participants: Normal control eyes and patients with age-related macular degeneration (AMD) with and without neMNV.

Methods: Swept-source OCT angiography (SS-OCTA) imaging (PLEX Elite 9000, Carl Zeiss Meditec, Inc) was performed using the 6 × 6-mm scan pattern. Individual B-scans were annotated to distinguish between drusen and the double-layer sign (DLS) associated with the neMNV. The machine learning model was tested on a dataset graded by humans, and model performance was compared with the human graders.

Main outcome measures: Intersection over Union (IoU) score was measured to evaluate segmentation network performance. Area under the receiver operating characteristic curve values, sensitivity, specificity, and positive predictive value (PPV) and negative predictive value (NPV) were measured to assess the performance of the final classification performance. Chance-corrected agreement between the algorithm and the human grader determinations was measured with Cohen's kappa.

Results: A total of 251 eyes from 210 patients, including 182 eyes with DLS and 115 eyes with drusen, were used for model training. Of 125 500 B-scans, 6879 B-scans were manually annotated. A vision transformer segmentation model was built to extract DLS and drusen from B-scans. The extracted prediction masks from all B-scans in a volume were projected to an en face image, and an eye-level projection map was obtained for each eye. A binary classification algorithm was established to identify eyes with neMNV from the projection map. The algorithm achieved 82%, 90%, 79%, and 91% sensitivity, specificity, PPV, and NPV, respectively, on a separate test set of 100 eyes that were evaluated by human graders in a previous study. The area under the curve value was calculated as 0.91 (95% confidence interval, 0.85-0.98). The results of the algorithm showed excellent agreement with the senior human grader (kappa = 0.83, P < 0.001) and moderate agreement with the junior grader consensus (kappa = 0.54, P < 0.001).

Conclusions: Our network (code is available at https://github.com/uw-biomedical-ml/double_layer_vit) was able to detect the presence of neMNV from structural B-scans alone by applying a purely transformer-based model.

Keywords: AI, artificial intelligence; AMD, age-related macular degeneration; Age-related macular degeneration; CNN, convolutional neural network; DLS, double-layer sign; Deep learning; GA, geographic atrophy; IoU, Intersection over Union; MNV, macular neovascularization; Macular neovascularization; NPV, negative predictive value; OCT; OCTA, OCT angiography; PPV, positive predictive value; ROC, receiver operating characteristic; RPE, retinal pigment epithelium; SS-OCT, swept-source OCT; SS-OCTA, swept-source OCT angiography; ViT, Vision Transformer; iAMD, intermediate age-related macular degeneration; neMNV, nonexudative macular neovascularization.