Correlation of vascular and fluid-related parameters in neovascular age-related macular degeneration using deep learning

Abstract

Purpose: To identify correlations between the vascular characteristics of macular neovascularization (MNV) obtained by optical coherence tomography angiography (OCTA) and distinct retinal fluid volumes in neovascular age-related macular degeneration (nAMD).

Methods: In this prospective interventional study, 54 patients with treatment-naïve type 1 or 2 nAMD were included and treated with intravitreal aflibercept. At baseline and month 1, each patient underwent a SD-OCT volume scan and volumetric flow scan using a swept-source OCTA. A deep learning algorithm was used to automatically detect and quantify fluid in OCT scans. Angio Tool, a National Cancer Institute algorithm, was used to skeletonize MNV properties and quantify lesion size (LS), vessel area (VA), vessel density (VD), total number of endpoints (TNE), total number of junctions (TNJ), junction density (JD), total vessel length (TVL), average vessel length (AVL) and mean-e-lacunarity (MEL). Subsequently, linear regression models were used to investigate a correlation between OCTA parameters and fluid quantifications.

Results: The median amount of fluid within the central 6-mm EDTRS ring was 173.7 nl at baseline, consisting of 156.6 nl of subretinal fluid (SRF) and 2.3 nl of intraretinal fluid (IRF). Fluid decreased significantly in all compartments to 1.76 nl (SRF) and 0.64 nl (IRF). The investigated MNV parameters did not change significantly after the first treatment. There was no significant correlation between MNV parameters and relative fluid decrease after anti-VEGF treatment. Baseline fluid correlated statistically significant but weakly with TNE (p = 0.002, R² = 0.17), SRF with TVL (p = 0.04, R² = 0.08), VD (p = 0.046, R² = 0.08), TNE (p = 0.001, R² = 0.20) and LS (p = 0.033, R² = 0.09). IRF correlated with VA (p = 0.042, R² = 0.08).The amount of IRF at month 1 correlated significantly but weakly with VD (p = 0.036, R² = 0.08), JD (p = 0.019, R² = 0.10) and MEL (p = 0.005, R² = 0.14).

Conclusion: Macular neovascularization parameters at baseline and month 1 played only a minor role in the exudation process in nAMD. None of the MNV parameters were correlated with the treatment response.

FIGURE 1

Image (a) and (d) show the en‐face scan of a type I MNV membrane at baseline (a) and 30 days (d) after the first anti‐VEGF intravitreal injection. Images (b) and (e) show skeletonized visualizations of the MNV membrane using the AngioTool software (red lines represent vessels, and blue dots junctions). Images (c) and (f) show the B scans with deep learning–based fluid detection at the levels of the red lines in images (a) and (d). Automatically detected intraretinal fluid is in orange‐yellow colour and subretinal fluid in blue.

FIGURE 2

Image (a) shows an OCTA enface image of one study patient at baseline with a mixed type I and 2 MNV and a visual acuity of 20/40. (b) shows the SLO image of the same patient, with AI‐based fluid detection. Blueish areas represent subretinal fluid (SRF), and yellowish areas represent intraretinal fluid (IRF). (c) shows the B scan at the level of the yellow central horizontal line in image (b). Blueish areas represent automatically detected and quantified SRF, and yellowish areas represent IRF. (d), (e) and (f) are the corresponding images 30 days after the first anti‐VEGF treatment.

FIGURE 3

Image (a) shows an OCTA enface image of one study patient at baseline with type I MNV and a visual acuity of 20/40. (b) shows the SLO image of the same patient, with AI‐based fluid detection. Blueish areas represent subretinal fluid (SRF), and yellowish areas represent intraretinal fluid (IRF). (c) shows the B scan at the level of the yellow central horizontal line in image (b). Blueish areas represent automatically detected and quantified SRF, and yellowish areas represent IRF. (d), (e) and (f) are the corresponding images 30 days after the first anti‐VEGF treatment.