Diabetic retinopathy is one of the leading causes of vision loss in the United States and\nother countries around the world. People who have diabetic retinopathy may not have symptoms\nuntil the condition becomes severe, which may eventually lead to vision loss. Thus, the medically\nunderserved populations are at an increased risk of diabetic retinopathy-related blindness. In this\npaper, we present development efforts on an embedded vision algorithm that can classify healthy\nversus diabetic retinopathic images. Convolution neural network and a k-fold cross-validation\nprocess were used. We used 88,000 labeled high-resolution retina images obtained from the publicly\navailable Kaggle/EyePacs database. The trained algorithm was able to detect diabetic retinopathy\nwith up to 76% accuracy. Although the accuracy needs to be further improved, the presented results\nrepresent a significant step forward in the direction of detecting diabetic retinopathy using embedded\ncomputer vision. This technology has the potential of being able to detect diabetic retinopathy\nwithout having to see an eye specialist in remote and medically underserved locations, which can\nhave significant implications in reducing diabetes-related vision losses.
Loading....