Prediction of Individual Disease Conversion in Early AMD Using Artificial Intelligence

Abstract

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.

Association between Drusen Burden Determined by OCT and Genetic Risk in Early and Intermediate Age-Related Macular Degeneration

“…48 Schmidt-Erfurth et al found that drusen area was an important quatitative feature for progression. 16 Higher drusen volume was related to advanced AMD in a retrospective review by Lei et al, but other OCT parameters were more strongly related in their cases. 23 A combination of drusen parameters along with other OCT derived parameters such as hyperreflective foci and retinal thickness may be the most informative and predictive of progression.…”

“…SD-OCT can quantify variation in morphology in a way that is accurate and reproducible. Other AMD phenotypes detectable on OCT including subfoveal fluid, retinal and choroidal thickness, retinal hyperreflective foci, other drusen characteristics such as hyper-reflectivity and homogeneity, are opportunities for exploration in the context of subtyping AMD and assessing risk of progression, 15,16,23,48 as well as defining the associations with genetic variants.…”

“…23 A combination of drusen parameters along with other OCT derived parameters such as hyperreflective foci and retinal thickness may be the most informative and predictive of progression. 15,16,23 The association between advanced AMD and the CFH Y402H variant as well as the intronic SNP in this gene, CFH rs1410996, have been well documented 6,20,34,38,[49][50][51] The biologic mechanisms underlying their effect on drusen accumulation, however, have not been fully explored. In our previous analyses of these variants regarding progression, the effect of rs1410996 was stronger and the effect of Y402H was reduced when both were in the same model, suggesting that this variant may be more strongly associated with AMD.…”

“…Prior studies have indicated that OCT-derived drusen area and volume, and retinal features such as hyper-reflective foci and choroidal parameters, are informative in determining the likelihood of progression from early or intermediate to advanced AMD. [11][12][13][14][15][16] However, knowledge about the association between OCT derived retina parameters and genetic variants related to AMD is sparse. [17][18][19] We previously reported the effects of genes on progression to different stages of AMD.…”

“…We evaluated the association between genetic risk and OCT drusen area and volume in a clinical cohort of patients with early and intermediate AMD. Understanding these relationships, in addition to other OCT parameters, [11][12][13][14][15][16]23 may lead to clinical use of these measurements with identification of earlier high risk phenotypes and better stratification of risk of progression. Further characterization of drusen morphology on OCT may lead to the identification of disease biomarkers that could serve as anatomic endpoints for clinical trials.…”