深度学习算法量化视盘照片中神经视网膜边缘丢失
本研究训练深度学习算法,从视盘照片预测SDOCT测量的BMO-MRW,以量化青光眼性神经视网膜损伤。共纳入927眼9282对图像,预测值与实测值高度相关(r=0.88),区分青光眼与正常眼的AUC为0.945,与SDOCT实测的0.933无显著差异(P=.587)。
要点
- 深度学习可从视盘照片准确预测SDOCT的BMO-MRW,平均绝对误差27.8μm。
- 预测BMO-MRW区分青光眼性视野缺损与正常眼的AUC达0.945。
- 该算法有望替代人工判读视盘照片,辅助青光眼检测。
核心结论
- 深度学习网络能以SDOCT的BMO-MRW为参考,从视盘照片量化神经视网膜损伤,并高精度检测青光眼。
研究设计与样本
- 横断面研究,纳入490名受试者927眼的9282对视盘照片与SDOCT视神经头扫描,随机分为训练+验证集(80%)和测试集(20%)。
主要发现
- 测试集中深度学习预测的全局BMO-MRW均值228.8±63.1μm与SDOCT实测226.0±73.8μm高度相关(Pearson r=0.88;R2=77%;P<.001),平均绝对误差27.8μm。
- 深度学习预测与SDOCT实测全局BMO-MRW区分青光眼与健康眼的AUC分别为0.945(95%CI 0.874-0.980)和0.933(95%CI 0.856-0.975),两者比较P=.587。
关键数值与推荐
- 深度学习预测BMO-MRW与SDOCT实测值高度相关(r=0.88),平均绝对误差27.8μm。
- 深度学习预测区分青光眼性视野丢失与正常眼的AUC为0.945(95%CI 0.874-0.980)。
临床怎么用
- 该算法可基于视盘照片量化神经视网膜边缘丢失,辅助青光眼检测,可能减少人工判读需求。
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 train a deep learning (DL) algorithm that quantifies glaucomatous neuroretinal damage on fundus photographs using the minimum rim width relative to Bruch membrane opening (BMO-MRW) from spectral-domain optical coherence tomography (SDOCT). DESIGN: Cross-sectional study. METHODS: A total of 9282 pairs of optic disc photographs and SDOCT optic nerve head scans from 927 eyes of 490 subjects were randomly divided into the validation plus training (80%) and test sets (20%). A DL convolutional neural network was trained to predict the SDOCT BMO-MRW global and sector values when evaluating optic disc photographs. The predictions of the DL network were compared to the actual SDOCT measurements. The area under the receiver operating curve (AUC) was used to evaluate the ability of the network to discriminate glaucomatous visual field loss from normal eyes. RESULTS: The DL predictions of global BMO-MRW from all optic disc photographs in the test set (mean ± standard deviation [SD]: 228.8 ± 63.1 μm) were highly correlated with the observed values from SDOCT (mean ± SD: 226.0 ± 73.8 μm) (Pearson's r = 0.88; R2 = 77%; P < .001), with mean absolute error of the predictions of 27.8 μm. The AUCs for discriminating glaucomatous from healthy eyes with the DL predictions and actual SDOCT global BMO-MRW measurements were 0.945 (95% confidence interval [CI]: 0.874-0.980) and 0.933 (95% CI: 0.856-0.975), respectively (P = .587). CONCLUSIONS: A DL network can be trained to quantify the amount of neuroretinal damage on optic disc photographs using SDOCT BMO-MRW as a reference. This algorithm showed high accuracy for glaucoma detection, and may potentially eliminate the need for human gradings of disc photographs.
引用
- Thompson AC, Jammal AA, Medeiros FA, et al. A Deep Learning Algorithm to Quantify Neuroretinal Rim Loss From Optic Disc Photographs. Am J Ophthalmol. 2019. PMID: 30689990. doi:10.1016/j.ajo.2019.01.011
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。