用普通相机拍眼睛前段，靠AI判断是否糖尿病，准确率超98%。
Automated Diabetic Screening via Anterior Segment Ocular Imaging: A Deep Learning and Explainable AI Approach
- 用眼表图像中的虹膜、巩膜等特征，结合深度学习识别糖尿病状态。
- 最佳模型在2640张图像上达98.21%的F1分数，正常人识别精准率100%。
- 适合基层医疗筛查，无需专业设备，且结果可解释，便于医生信任。
糖尿病视网膜病变筛查传统依赖眼底摄影，需专业设备与人员，难以在基层及资源匮乏地区普及。本文开发并验证了一种基于深度学习（DL）的自动化糖尿病分类系统，利用常规摄影设备获取的眼前段影像作为替代方案。该系统挖掘虹膜、巩膜和结膜中与全身糖尿病状态相关的可见生物标志物。我们在2,640张临床标注的面前段图像上，系统评估了五种主流架构（EfficientNet-V2-S带自监督学习（SSL）、Vision Transformer、Swin Transformer、ConvNeXt-Base、ResNet-50），涵盖正常、可控糖尿病、不可控糖尿病三类。采用融合镜面反射抑制与对比度受限自适应直方图均衡化（CLAHE）的预处理流程，以增强关键血管与纹理模式。在领域特定眼部图像上使用SimCLR进行自监督学习显著提升模型性能。EfficientNet-V2-S结合SSL达到最优表现：F1分数98.21%，精确率97.90%，召回率98.55%，远超仅使用ImageNet初始化的模型（F1 94.63%）。尤为突出的是，正常人群分类精确率达到近乎完美的100%，对减少不必要的临床转诊至关重要。
原文摘要 · Abstract (English)
Diabetic retinopathy screening traditionally relies on fundus photography, requiring specialized equipment and expertise often unavailable in primary care and resource limited settings. We developed and validated a deep learning (DL) system for automated diabetic classification using anterior segment ocular imaging a readily accessible alternative utilizing standard photography equipment. The system leverages visible biomarkers in the iris, sclera, and conjunctiva that correlate with systemic diabetic status. We systematically evaluated five contemporary architectures (EfficientNet-V2-S with self-supervised learning (SSL), Vision Transformer, Swin Transformer, ConvNeXt-Base, and ResNet-50) on 2,640 clinically annotated anterior segment images spanning Normal, Controlled Diabetic, and Uncontrolled Diabetic categories. A tailored preprocessing pipeline combining specular reflection mitigation and contrast limited adaptive histogram equalization (CLAHE) was implemented to enhance subtle vascular and textural patterns critical for classification. SSL using SimCLR on domain specific ocular images substantially improved model performance.EfficientNet-V2-S with SSL achieved optimal performance with an F1-score of 98.21%, precision of 97.90%, and recall of 98.55% a substantial improvement over ImageNet only initialization (94.63% F1). Notably, the model attained near perfect precision (100%) for Normal classification, critical for minimizing unnecessary clinical referrals.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。