A prospective clinical trial has been successfully completed for the detection of diabetic retinopathy using EyeCheckup. In this study 3 different models of fundus cameras were validated on 900 volunteer diabetic patients at Akdeniz University with an average accuracy of over 96%.
Fundus images of the posterior region of the eye were taken from voluntary patients using non-mydriatic fundus cameras (Canon, Topcon, and Optomed).
These photos were evaluated and graded for the presence and different severity levels of diabetic retinopathy and diabetic macular edema using ICDR scale. The human grading output was compared to EyeCheckup and EyeCheckup was found to be highly accurate.