Researchers from the Medical University of Vienna and Genetech studied 495 eyes with intermediate AMD, choiroidal neovascularization (CNV) or geographic atrophy (GA). They reviewed automated volumetric segmentation of outer neurosensory layers and retinal pigment epithelium, drusen and hyper-reflective foci with spectral domain-optical coherence tomography. By using imaging in conjunction with demographic and genetic input, they developed a predictive model that assessed the risk of developing advanced AMD.

Researchers found 159 eyes (32%) had developed advanced AMD within two years, while 114 eyes progressed to CNV and 45 to GA. The most critical quantitative features for progression, they found, were outer retinal thickness, hyper-reflective foci and drusen area. The features for conversion showed pathognomonic patterns that were distinctly different for the neovascular and atrophic pathways. Predictive hallmarks for CNV were mostly drusen-centric, while GA markers were associated with neurosensory retina and age, the study reported.

“Artificial intelligence with automated analysis of imaging biomarkers allows personalized prediction of AMD progression. Moreover, pathways of progression may be specific in respect to the neovascular/atrophic type,” the researchers said.

Schmidt-Erfurth U, Waldstein SM, Klimscha S, et al. Prediction of individual disease conversion in early AMD using artificial intelligence. Invest Ophthalmol Vis Sci. 2018 Jul 2;59(8):3199-08.