Ganjdanesh Alireza, Zhang Jipeng, Chew Emily Y, Ding Ying, Huang Heng, Chen Wei
Department of Electrical and Computer Engineering, Swanson School of Engineering, University of Pittsburgh, Pittsburgh, PA 15261, USA.
Department of Biostatistics, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA 15213, USA.
PNAS Nexus. 2022 Mar 19;1(1):pgab003. doi: 10.1093/pnasnexus/pgab003. eCollection 2022 Mar.
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.
年龄相关性黄斑变性（AMD）是发达国家失明的主要原因，到2040年其患病率将增至2.88亿人。因此，自动分级和预测方法对于识别晚期AMD易感人群并使临床医生能够为他们开展预防措施具有极大益处。临床上，AMD严重程度通过视网膜彩色眼底照片（CFP）进行量化，并且已经提出了许多基于机器学习的方法来对AMD严重程度进行分级。然而，很少有模型用于预测纵向进展情况，即根据当前的CFP预测未来晚期AMD风险，而这在临床上更具意义。在本文中，我们提出了一种新的基于深度学习的分类模型（LONGL-Net），它可以同时对当前CFP进行分级并预测纵向结果，即该受试者在未来时间点是否会发展为晚期AMD。我们设计了一种新的基于时间相关性结构引导的生成对抗网络模型，该模型学习连续时间点CFP中时间变化的相互关系，并通过预测未来CFP中的AMD症状为分类器的决策提供可解释性。我们使用了年龄相关性眼病研究中4628名参与者的约30000张CFP图像。在同时对当前时间点的AMD状况进行分级并预测受试者未来时间点晚期AMD进展的3类分类问题上，我们的分类器的曲线下面积（AUC）平均为0.905（95%置信区间：0.886 - 0.922），准确率为0.762（95%置信区间：0.733 - 0.792）。我们在英国生物银行数据集上进一步验证了我们的模型，在对300张CFP图像进行分级时，我们的模型准确率平均为0.905，灵敏度为0.797。
