Analysing the Progression Rates of Macular Lesions with Autofluorescence Imaging Modes in Dry Age-Related Macular Degeneration

AI-driven generalized polynomial transformation models for unsupervised fundus image registration

We introduce a novel AI-driven approach to unsupervised fundus image registration utilizing our Generalized Polynomial Transformation (GPT) model. Through the GPT, we establish a foundational model capable of simulating diverse polynomial transformations, trained on a large synthetic dataset to encompass a broad range of transformation scenarios. Additionally, our hybrid pre-processing strategy aims to streamline the learning process by offering model-focused input. We evaluated our model's effectiveness on the publicly available AREDS dataset by using standard metrics such as image-level and parameter-level analyzes. Linear regression analysis reveals an average Pearson correlation coefficient (R) of 0.9876 across all quadratic transformation parameters. Image-level evaluation, comprising qualitative and quantitative analyzes, showcases significant improvements in Structural Similarity Index (SSIM) and Normalized Cross Correlation (NCC) scores, indicating its robust performance. Notably, precise matching of the optic disc and vessel locations with minimal global distortion are observed. These findings underscore the potential of GPT-based approaches in image registration methodologies, promising advancements in diagnosis, treatment planning, and disease monitoring in ophthalmology and beyond.

The Use of Fundus Autofluorescence in Dry Age-Related Macular Degeneration

Fundus autofluorescence (FAF) has been a well-known imaging method for quite some time. However, with developing technologies and novel imaging devices, FAF is being used more often to diagnose and monitor retinal diseases. The density of lipofuscin (LF) and other fluorophores in the retina have a determining role in FAF images. In dry age-related macular degeneration (AMD), hyperautofluorescence is seen in cases of increasing LF in the retina pigment epithelium, whereas hypoautofluorescence is detected in decreasing LF resulting from geographic atrophy. In recent years, studies have shown that FAF images provide prognostic information in patients with AMD. This review aims to highlight the importance of FAF imaging in dry AMD.

Impact of ocular and systemic risk factors on progression of geographic atrophy in age-related macular degeneration

BACKGROUND

Geographic atrophy (GA) corresponds to an end-stage form of age-related macular degeneration and the leading cause of irreversible vision loss. The aim of this study was to investigate the role of fundus autofluorescence (FAF) patterns surrounding GA, baseline GA size, the presence of reticular pseudodrusen (RPD), the condition of the fellow eye and the presence of systemic vascular diseases (diabetes, atherosclerosis, systemic hypertension) on the progression of GA due to age-related macular degeneration.

METHODS

Seventy-five eyes of 52 patients with GA were included into this retrospective study. Digital FAF images were achieved using scanning laser ophthalmoscopy. Areas of GA were measured with the Region Finder software program on an Heidelberg Retinal Angiograph 2. FAF patterns around GA were classified and the presence of RPD and systemic diseases were noted. The median follow-up of the patients was 2.6 years (interquartile range, 1-9.2).

RESULTS

The median progression rate of the banded pattern (median 0.97 mm²/year) was significantly higher than the other patterns (median 0.85 mm²/year) (p = 0.03). Eyes with RPD had a significantly higher progression rate (median 1.21 mm²/year) than those without RPD (median 0.79 mm²/year) (p = 0.007). The presence of systemic diseases was not related with lesion progression rates.

CONCLUSIONS

The results demonstrate that the banded pattern and the presence of RPD accelerate the progression of GA and function as predictive factors.