LONGL-Net: temporal correlation structure guided deep learning model to predict longitudinal age-related macular degeneration severity

Abstract

Age-related macular degeneration (AMD) is the principal cause of blindness in developed countries, and its prevalence will increase to 288 million people in 2040. Therefore, automated grading and prediction methods can be highly beneficial for recognizing susceptible subjects to late-AMD and enabling clinicians to start preventive actions for them. Clinically, AMD severity is quantified by Color Fundus Photographs (CFP) of the retina, and many machine-learning-based methods are proposed for grading AMD severity. However, few models were developed to predict the longitudinal progression status, i.e. predicting future late-AMD risk based on the current CFP, which is more clinically interesting. In this paper, we propose a new deep-learning-based classification model (LONGL-Net) that can simultaneously grade the current CFP and predict the longitudinal outcome, i.e. whether the subject will be in late-AMD in the future time-point. We design a new temporal-correlation-structure-guided Generative Adversarial Network model that learns the interrelations of temporal changes in CFPs in consecutive time-points and provides interpretability for the classifier's decisions by forecasting AMD symptoms in the future CFPs. We used about 30,000 CFP images from 4,628 participants in the Age-Related Eye Disease Study. Our classifier showed average 0.905 (95% CI: 0.886–0.922) AUC and 0.762 (95% CI: 0.733–0.792) accuracy on the 3-class classification problem of simultaneously grading current time-point's AMD condition and predicting late AMD progression of subjects in the future time-point. We further validated our model on the UK Biobank dataset, where our model showed average 0.905 accuracy and 0.797 sensitivity in grading 300 CFP images.

Predictive Deep Learning Applications in Ophthalmology

“…that learned from temporal, longitudinal changes in CFPs; this model achieved an accuracy of 0.762 (95% CI 0.733-0.792) in simultaneously grading disease severity at the current time point and in predicting progression to late AMD at a future time point. 45 Lastly, Liefers et al…”

“…A number of studies used CFPs from the Age-Related Eye Disease Study (AREDS) clinical trial to train their neural networks. [41][42][43][44][45] Bhujyan et al applied a two-step approach: (1) classification of images according to disease severity using DL and (2) prediction of progression to advanced AMD using classical ML. 43 The hybrid model achieved 84% accuracy in predicting advanced AMD development (GA and nAMD) within 2 years.…”

Research progress on diagnosing retinal vascular diseases based on artificial intelligence and fundus images

“…In addition to extensive research on the diagnosis and classification of AMD, AI has been used to predict the severity, disease progression, and therapeutic effect in patients with age-related macular degeneration. Ganjdanesh et al (2022) created a new DL model (LONGL-Net) based on ResNet-18 to predict the severity and progression of patients with age-related macular degeneration. They collected approximately 30,000 color fundus photographs for training and verifying the model.…”

Joint Representation Learning with Generative Adversarial Imputation Network for Improved Classification of Longitudinal Data

“…Recent advancements have led to the consideration of longitudinal data components as distinct modalities for multimodal representation learning [17,18]. Research has shown that fusion strategies-combining features from multiple modalities into a joint representation-on longitudinal data components can improve various tasks such as sample generation [19], classification or prediction [20][21][22] and risk analysis [23].…”