Role of Artificial Intelligence in Retinal Diseases [German: Rolle der künstlichen Intelligenz bei verschiedenen retinalen Erkrankungen]

AI Generated Summary*

A review surveying how artificial intelligence, particularly deep learning on colour fundus photographs and OCT, is applied in retinal disease for screening, staging, lesion quantification, progression prediction and decision support. An FDA-approved diabetic retinopathy screening system showed 87% sensitivity and 91% specificity; OCT-based AMD staging reached an AUC of 0.94; conversion prediction achieved 68% accuracy for neovascular AMD and 80% for geographic atrophy; anti-VEGF need prediction reached AUC 70 to 77%. Automated segmentation of geographic atrophy trial OCT data indicated ellipsoid zone loss exceeds and precedes RPE loss. The central argument is that AI can process imaging data beyond human capacity, identify biomarkers and support individualised treatment decisions; two MDR-certified fluid and atrophy quantification tools are described.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.