Deep Learning for the Prediction of Uncorrected Refractive Error using OCT

Description

To develope a deep learning model for predicting uncorrected refractive (spherical equivalent, SE) error and keratometry (K value) using posterior (retinal) optical coherence tomography (OCT) images of the macula and optic nerve head. These materials are provided to help demonstrate and understand the concept of our research paper. The materials include the lightweight versions of the developed deep learning models (for Keras) and the codes for implementation.

Institutions

• Yonsei University

Categories

Licence