文献速递青光眼神经眼科

基于深度学习预测正常眼压性青光眼可疑者的青光眼转化

Deep-learning-based prediction of glaucoma conversion in normotensive glaucoma suspects.
PubMed 文献 · Br J Ophthalmol · 2024-06-20

本研究纳入随访至少7年且眼压低于21 mm Hg的210只正常眼压性青光眼可疑眼(105只转化为正常眼压性青光眼,105只未转化),利用眼底图像和临床特征训练三种机器学习分类器预测转化。三种算法均获得高诊断准确度,AUC范围为0.987至0.994;XGBoost对转化时间的预测表现最佳(均方误差2.24)。

要点

  • 三种机器学习分类器预测正常眼压性青光眼转化的AUC范围为0.987至0.994。
  • XGBoost对转化时间的预测表现最佳,均方误差为2.24。
  • 预测转化时间最重要的临床特征为基线眼压、舒张压和平均环视盘视网膜神经纤维层厚度。

核心结论

  • 使用眼底图像和临床数据训练的深度学习模型显示出预测正常眼压性青光眼可疑者是否及何时转化为正常眼压性青光眼的潜力。

研究设计与样本

  • 回顾了12 458只青光眼可疑眼的资料,纳入210只眼(105只转化为正常眼压性青光眼,105只未转化),随访至少7年且期间眼压低于21 mm Hg。
  • 通过卷积自编码器提取视盘照相和無赤光视网膜神经纤维层照相两种眼底图像的特征。
  • 将提取的图像特征与15项临床特征(包括年龄、性别、眼压、等效球镜度、中央角膜厚度、眼轴长度、平均环视盘视网膜神经纤维层厚度、收缩压/舒张压和体重指数)结合,使用XGBoost、随机森林和梯度提升三种机器学习分类器以不同特征组合进行预测。

主要发现

  • 三种算法对正常眼压性青光眼转化预测均达到高诊断准确度,AUC范围为0.987(95%CI 0.978至1.000;随机森林使用眼底图像和临床特征训练)至0.994(95%CI 0.984至1.000;XGBoost使用眼底图像和临床特征训练)。
  • XGBoost对转化时间的预测表现最佳,均方误差为2.24。
  • 预测转化时间最重要的三项临床特征为基线眼压、舒张压和平均环视盘视网膜神经纤维层厚度。

局限性

  • 原文未报告局限性。

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

BACKGROUND/AIMS: To assess the performance of deep-learning (DL) models for prediction of conversion to normal-tension glaucoma (NTG) in normotensive glaucoma suspect (GS) patients. METHODS: Datasets of 12 458 GS eyes were reviewed. Two hundred and ten eyes (105 eyes showing NTG conversion and 105 without conversion), followed up for a minimum of 7 years during which intraocular pressure (IOP) was lower than 21 mm Hg, were included. The features of two fundus images (optic disc photography and red-free retinal nerve fibre layer (RNFL) photography) were extracted by convolutional auto encoder. The extracted features as well as 15 clinical features including age, sex, IOP, spherical equivalent, central corneal thickness, axial length, average circumpapillary RNFL thickness, systolic/diastolic blood pressure and body mass index were used to predict NTG conversion. Prediction was performed using three machine-learning classifiers (ie, XGBoost, Random Forest, Gradient Boosting) with different feature combinations. RESULTS: All three algorithms achieved high diagnostic accuracy for NTG conversion prediction. The AUCs ranged from 0.987 (95% CI 0.978 to 1.000; Random Forest trained with both fundus images and clinical features) and 0.994 (95% CI 0.984 to 1.000; XGBoost trained with both fundus images and clinical features). XGBoost showed the best prediction performance for time to NTG conversion (mean squared error, 2.24). The top three important clinical features for time-to-conversion prediction were baseline IOP, diastolic blood pressure and average circumpapillary RNFL thickness. CONCLUSION: DL models, trained with both fundus images and clinical data, showed the potential to predict whether and when normotensive GS patients will show conversion to NTG.

来源:Br J Ophthalmol,查看原文。版权归原作者所有。

引用

  1. Ha A, Sun S, Kim YK, et al. Deep-learning-based prediction of glaucoma conversion in normotensive glaucoma suspects. Br J Ophthalmol. 2024. PMID: 37918891. doi:10.1136/bjo-2022-323167

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。