Multimodal imaging and deep learning in geographic atrophy secondary to age-related macular degeneration

Abstract

Geographic atrophy (GA) secondary to age-related macular degeneration is among the most common causes of irreversible vision loss in industrialized countries. Recently, two therapies have been approved by the US FDA. However, given the nature of their treatment effect, which primarily involves a relative decrease in disease progression, discerning the individual treatment response at the individual level may not be readily apparent. Thus, clinical decision-making may have to rely on the quantification of the slope of GA progression before and during treatment. A panel of imaging modalities and artificial intelligence (AI)-based algorithms are available for such quantifications. This article aims to provide a comprehensive overview of the fundamentals of GA imaging, the procedures for diagnosis and classification using these images, and the cutting-edge role of AI algorithms in automatically deriving diagnostic and prognostic insights from imaging data.

Keywords: age-related macular degeneration; artificial intelligence; geographic atrophy; imaging biomarkers; multimodal imaging.

Figure 1.. En-face Imaging of Geographic Atrophy

The panel shows a blue-excitation fundus autofluorescence (FAF), green-excitation FAF, and near-infrared reflectance (NIR) confocal scanning laser ophthalmoscopy images of a patient with geographic atrophy secondary to age-related macular degeneration. The green FAF image provides a more distinct boundary in the foveal region since it is unaffected by macular pigment. The lower right image is a spectral domain optical coherence tomography (SD-OCT) en-face projection of the mean signal posterior to the retinal pigment epithelium. Due to signal ‘hypertransmission’, the area of atrophy is evident.

Figure 2.. Artificial Intelligence-Based Quantification of Geographic Atrophy

The figure shows an artificial intelligence (AI)-based segmentation of a spectral domain optical coherence tomography (SD-OCT) B-scan. The lower left image shows the geographic atrophy (GA) segmentation on an en-face projection. The lower right image shows the standardized outer nuclear layer (ONL) thickness map in reference to the normal ONL thickness. Blue colors denote thinning by over two standard deviations below the normative mean. Of note, the ONL shows thinning within areas of atrophy and outside of the GA lesions. Abbreviations: retinal pigment epithelium drusen complex (RPEDC), outer nuclear layer (ONL), spectral domain optical coherence tomography (SD-OCT)