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

Abstract

Accurately predicting a patient's risk of progressing to late age-related macular degeneration (AMD) is difficult but crucial for personalized medicine. While existing risk prediction models for progression to late AMD are useful for triaging patients, none utilizes longitudinal color fundus photographs (CFPs) in a patient's history to estimate the risk of late AMD in a given subsequent time interval. In this work, we seek to evaluate how deep neural networks capture the sequential information in longitudinal CFPs and improve the prediction of 2-year and 5-year risk of progression to late AMD. Specifically, we proposed two deep learning models, CNN-LSTM and CNN-Transformer, which use a Long-Short Term Memory (LSTM) and a Transformer, respectively with convolutional neural networks (CNN), to capture the sequential information in longitudinal CFPs. We evaluated our models in comparison to baselines on the Age-Related Eye Disease Study, one of the largest longitudinal AMD cohorts with CFPs. The proposed models outperformed the baseline models that utilized only single-visit CFPs to predict the risk of late AMD (0.879 vs 0.868 in AUC for 2-year prediction, and 0.879 vs 0.862 for 5-year prediction). Further experiments showed that utilizing longitudinal CFPs over a longer time period was helpful for deep learning models to predict the risk of late AMD. We made the source code available at https://github.com/bionlplab/AMD_prognosis_mlmi2022 to catalyze future works that seek to develop deep learning models for late AMD prediction.

Keywords: Age-related macular degeneration; Convolutional neural networks; Deep learning; Recurrent neural networks; Transformer.

Fig. 1.

Examples of 2-year and 5-year late AMD prediction task. Each CFP in individual’s history is labeled 1 if late AMD onset is detected within a given prediction window (2 years and 5 years), otherwise 0. The blue lines represents the observation interval (t₀, t_l], the red line represents the prediction window (t_l, t_l + n]. A. Prediction scenario for 2 and 5 years. B. Training scenario for a single unrolled patient.

Fig. 2.

Model architectures. A. CNN-LSTM. B. CNN-Transformer.