In this project, we used machine learning to diagnose retinal and optic nerve diseases, including Choroidal Neovascularization, Diabetic Macular Edema, and Drusen, using the OCT-MNIST dataset of 109,309 images. We created two models: a logistic regression + Random Forest pipeline and a CNN. The CNN model showed superior performance.
The dataset used in this project can be obtained from the original source: https://doi.org/10.1016/j.cell.2018.02.010
Please follow the terms and conditions specified by the authors before downloading and using the dataset.