Predicting Age-related Macular Degeneration Progression with Longitudinal Fundus Images using Deep Learning

This repository contains source codes for implementing models and reproducing the results in the following paper:

Predicting Age-related Macular Degeneration Progression with Longitudinal Fundus Images using Deep Learning
    Junghwan Lee, Tingyi Wanyan, Qingyu Chen, Tiarnan D.L. Keenan, Benjamin S. Glicksberg, Emily Y. Chew, Zhiyong Lu, Fei Wang, and Yifan Peng*
    Machine Learning in Medical Imaging 2022, https://link.springer.com/chapter/10.1007/978-3-031-21014-3_2

Installation

The source codes were written in Python 3.7.1 with Tensorflow version above 2.5.0

Data access

We used the data from AREDS, which was sponsored by the National Eye Institute of the National Institutes of Health. The data is publicly available upon request: https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000001.v3.p1.

To implement the model

Source codes for all models are available in src folder. Please refer model_main.py for implementation and hyperparameter setting.

Supplementary material for the paper

Please send any questions about the code to jlee3541 at gatech.edu.