# Improving OCTA Visualization of Macular Neovascularization via a Grayscale Inversion Method

## TL;DR

This paper introduces a grayscale inversion method to improve the visibility of macular neovascularization in OCT angiography images, potentially aiding clinical assessments.

## Contribution

The study proposes a simple grayscale inversion technique using ImageJ to enhance visualization of macular neovascularization in OCTA.

## Key findings

- Grayscale inversion significantly improved visualization of macular neovascularization in 6 × 6 mm and 12 × 12 mm OCTA scans.
- The method also improved detection of polypoidal lesions in OCTA images.
- No significant improvement was observed in 3 × 3 mm scans or area-related features.

## Abstract

Background: Age-related macular degeneration is a major cause of vision loss, and improved visualization of macular neovascularization (MNV) on OCT angiography (OCTA) could enhance clinical assessment. This study aimed to establish a simple and accessible image enhancement method. Methods: We retrospectively analyzed 24 eyes from 22 patients with MNV at the Doheny UCLA Eye Centers. Grayscale-inverted OCTA images were generated using the basic “Invert” function in ImageJ 1.51 23. Each original and inverted image pair was assessed for seven MNV-related features: structure and area within 3 × 3 mm, 6 × 6 mm and 12 × 12 mm scans, and presence of polypoidal lesions. Twenty-one ophthalmologists graded visibility using a standardized five-point scale. Paired comparisons were performed using the Wilcoxon signed-rank test. Results: Grayscale inversion significantly improved the visualization of MNV structure in 6 × 6 mm scans (mean difference: +0.67 ± 1.02; p = 0.008), 12 × 12 mm scans (+0.62 ± 1.07; p = 0.013), and detection of polypoidal lesions (+0.43 ± 0.98; p = 0.030). No significant differences were found for 3 × 3 mm structure (p = 0.793) or area-related features (all p > 0.3). Conclusions: Grayscale inversion may enhance MNV visibility and polypoidal lesion detection on OCTA. As this study relied solely on subjective assessments, future work should incorporate quantitative image analysis.