AI‐based studies using fundus images

Abstract

The artificial intelligence (AI)/deep learning (DL) algorithms were shown to detect diseases based on image analysis, including retinal diseases from fundus photos and optical coherence tomography (OCT) scans, lung diseases from chest radiographs, and skin disorders from skin photos. It was also shown that retinal photos can be used to identify risk factors related with cardiovascular disorders, including blood pressure, smoking and body mass index.The fundus picture was shown to source data for developing AI/DL algorithms detecting many ophthalmic diseases and pathologies, including diabetic retinopathy, diabetic macular edema, retinal artery and vein occlusion, macular hole, epiretinal membrane, AMD, glaucoma, retinopathy of prematurity, myopia, myopic macular degeneration, hyperopia, optic disc edema, optic nerve atrophy, central serous retinopathy, myelinated nerve fibres, and retinitis pigmentosa. Moreover, fundus pictures were also used to develop many AI/DL algorithms detecting non ophthalmic diseases and pathologies, including Alzheimer and Parkinson disease, risk of stroke, risk of cardiac infarction, systemic hypertension, chronic renal disease, peripheral arterial disease, and anaemia.