Longitudinal imaging is able to capture both static anatomical structures and dynamic changes in disease progression towards earlier and better patient-specific pathology management. However, conventional approaches for detecting diabetic retinopathy (DR) rarely take advantage of longitudinal information to improve DR analysis. In this work, we investigate the benefit of exploiting self-supervised learning with a longitudinal nature for DR diagnosis purposes. We compare different longitudinal self-supervised learning (LSSL) methods to model the disease progression from longitudinal retinal color fundus photographs (CFP) to detect early DR severity changes using a pair of consecutive exams. The experiments were conducted on a longitudinal DR screening dataset with or without those trained encoders (LSSL) acting as a longitudinal pretext task. Results achieve an AUC of 0.875 for the baseline (model trained from scratch) and an AUC of 0.96 (95% CI: 0.9593-0.9655 DeLong test) with a p-value < 2.2e-16 on early fusion using a simple ResNet alike architecture with frozen LSSL weights, suggesting that the LSSL latent space enables to encode the dynamic of DR progression.
翻译:纵向成像能够捕捉静态解剖结构和疾病向早期和更好的病人病理管理进展的动态变化,然而,发现糖尿病视网膜病(DR)的常规方法很少利用纵向信息来改进DR分析。在这项工作中,我们调查为DR诊断目的利用纵向性质自我监督学习的好处。我们比较了不同的纵向自我监督学习(LSSL)方法,以模拟疾病从纵向视网膜花基照片(CFP)发展到利用连续的一对测试来检测早期DR严重程度变化。实验是在长视DR筛选数据集上进行的,无论是否由受过训练的编码员(LSSLSL)作为长视托任务。结果为基线(从头到脚的模型)和0.96(95%CI:0.5993-0.9655 DeLong测试)提供了0.75AUC(95% CI:0.9593-09655 DeL)和P-2.2e-16关于利用简单的ResNet等结构进行早期融合的P-2.2E-16模型,表明LSS潜层空间代码能够进行动态发展。