1. Neural Networks to Classify Diabetic Retinopathy
We used Residual Networks coupled with the cyclic pooling technique for data refinement to train on 35,126 retinal images. ... A support vector machine (SVM), a type of binary classifier related to neural networks that attempt to construct a hyperplane with the widest margin between data points [1], has previously been used to classify non-proliferative DR from proliferative DR to a high accuracy [1]. ... To diagnose the retinal images, we developed a neural network architecture centered on the current state-of-the-art convolutional neural network for classifying images -- Residual Networks (R...
- Word Count: 874
- Approx Pages: 3
- Grade Level: Graduate