Diabetic Retinopathy Detection Using 3D Oct Features

If untreated, diabetic retinopathy (DR), a serious health issue, can cause visual loss. In this paper, a computer-aided diagnostic (CAD) system for DR detection is developed using 3D optical coherence tomography (OCT) images. The retinal layers are first separated from the input 3D OCT images. Second, the first-order reflectivity and the thickness of each OCT layer are computed as 3D characteristics from each retinal layer. Finally, backpropagation neural networks are used for classification. Utilizing {10-folds} cross-validation, experiments on 188 cases confirm the benefits of the proposed system over competing approaches, with an accuracy of 94.74%+-5.55%. These results demonstrate the method's potential for DR detection utilizing OCT images.