深度学习区分活动性角膜感染与愈合瘢痕
该研究训练卷积神经网络(CNN),仅用外部照片区分活动性角膜溃疡与愈合瘢痕。在印度和北加州两个独立人群测试,F1分数分别为92.0%和84.3%,可视化相关特征包括角膜浸润、前房积脓和结膜充血。
要点
- CNN仅用外部照片区分角膜溃疡与瘢痕,印度人群F1为92.0%,北加州为84.3%。
- 模型可视化关注角膜浸润、前房积脓和结膜充血等临床相关特征。
- 有望作为低成本诊断工具,辅助眼保健资源有限地区的分诊。
核心结论
- CNN能以高准确度分类角膜溃疡与瘢痕,并可泛化到训练数据之外的患者人群。
- CNN诊断时聚焦于临床相关特征,有望作为低成本诊断方法辅助分诊。
研究设计与样本
- 研究为卷积神经网络训练与测试,使用角膜溃疡和瘢痕的去标识照片。
- 训练集来自SCUT和MUTT,溃疡照片n=1313、瘢痕照片n=1132。
- 测试集来自印度眼科诊所(n=200)和斯坦福Byers眼科研究所(n=101)两个不同患者人群。
- 以金标准临床分类评估准确度,主要指标为F1分数,并用ROC曲线下面积(AUC)衡量精确率-召回率权衡。
主要发现
- 印度角膜溃疡患者中,CNN正确分类123例活动性溃疡中的115例和77例瘢痕中的65例。
- 印度人群F1分数92.0%(95%CI 88.2%-95.8%),敏感度93.5%(95%CI 89.1%-97.9%),特异度84.42%(95%CI 79.42%-89.42%),ROC AUC 0.9731。
- 北加州角膜溃疡患者中,CNN正确分类55例活动性溃疡中的43例和46例瘢痕中的42例。
- 北加州人群F1分数84.3%(95%CI 77.2%-91.4%),敏感度78.2%(95%CI 67.3%-89.1%),特异度91.3%(95%CI 85.8%-96.8%),ROC AUC 0.9474。
- CNN可视化与角膜浸润、前房积脓和结膜充血等临床相关特征相关。
临床怎么用
- 可考虑将CNN作为仅用外部照片的低成本辅助工具,用于活动性角膜溃疡与愈合瘢痕的鉴别。
- 在眼保健资源有限地区,该算法可能辅助分诊,但需结合临床评估。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
PURPOSE: To develop and evaluate an automated, portable algorithm to differentiate active corneal ulcers from healed scars using only external photographs. DESIGN: A convolutional neural network was trained and tested using photographs of corneal ulcers and scars. PARTICIPANTS: De-identified photographs of corneal ulcers were obtained from the Steroids for Corneal Ulcers Trial (SCUT), Mycotic Ulcer Treatment Trial (MUTT), and Byers Eye Institute at Stanford University. METHODS: Photographs of corneal ulcers (n = 1313) and scars (n = 1132) from the SCUT and MUTT were used to train a convolutional neural network (CNN). The CNN was tested on 2 different patient populations from eye clinics in India (n = 200) and the Byers Eye Institute at Stanford University (n = 101). Accuracy was evaluated against gold standard clinical classifications. Feature importances for the trained model were visualized using gradient-weighted class activation mapping. MAIN OUTCOME MEASURES: Accuracy of the CNN was assessed via F1 score. The area under the receiver operating characteristic (ROC) curve (AUC) was used to measure the precision-recall trade-off. RESULTS: The CNN correctly classified 115 of 123 active ulcers and 65 of 77 scars in patients with corneal ulcer from India (F1 score, 92.0% [95% confidence interval (CI), 88.2%-95.8%]; sensitivity, 93.5% [95% CI, 89.1%-97.9%]; specificity, 84.42% [95% CI, 79.42%-89.42%]; ROC: AUC, 0.9731). The CNN correctly classified 43 of 55 active ulcers and 42 of 46 scars in patients with corneal ulcers from Northern California (F1 score, 84.3% [95% CI, 77.2%-91.4%]; sensitivity, 78.2% [95% CI, 67.3%-89.1%]; specificity, 91.3% [95% CI, 85.8%-96.8%]; ROC: AUC, 0.9474). The CNN visualizations correlated with clinically relevant features such as corneal infiltrate, hypopyon, and conjunctival injection. CONCLUSIONS: The CNN classified corneal ulcers and scars with high accuracy and generalized to patient populations outside of its training data. The CNN focused on clinically relevant features when it made a diagnosis. The CNN demonstrated potential as an inexpensive diagnostic approach that may aid triage in communities with limited access to eye care.
引用
- Tiwari M, Piech C, Baitemirova M, et al. Differentiation of Active Corneal Infections from Healed Scars Using Deep Learning. Ophthalmology. 2021. PMID: 34352302. doi:10.1016/j.ophtha.2021.07.033
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。