文献速递青光眼眼底病与葡萄膜炎

基于黄斑OCT厚度图深度学习估算10-2与24-2视野指标

Deep Learning Estimation of 10-2 and 24-2 Visual Field Metrics Based on Thickness Maps from Macula OCT.
PubMed 文献 · Ophthalmology · 2021-04-23

本研究利用645名健康与青光眼受试者(1222眼)的2408对10-2视野-SD OCT及2999对24-2视野-SD OCT数据,训练深度学习模型从黄斑SD OCT厚度图估算视野平均偏差(MD)与模式标准差(PSD)。联合模型估算10-2 MD的R2为0.82、MAE为1.9 dB,估算24-2 MD的R2为0.79、MAE为2.1 dB,显著优于平均厚度线性回归。

要点

  • 联合深度学习模型估算10-2 MD的R2为0.82,MAE为1.9 dB。
  • 联合模型估算24-2 MD的R2为0.79,MAE为2.1 dB,优于平均厚度回归。
  • GCIPL与GCC分别在估算10-2与24-2 MD时表现最佳。

核心结论

  • 深度学习模型可从SD OCT图像改善功能损失的估算,有助于个体化视野检查。

研究设计与样本

  • 诊断技术评估,纳入645名健康与青光眼受试者共1222眼,收集2408对10-2视野-SD OCT及2999对24-2视野-SD OCT数据。
  • 基于Spectralis黄斑SD OCT的6层厚度图训练深度学习模型,并与平均层厚度线性回归比较。

主要发现

  • 联合模型估算10-2 MD的R2为0.82(95%CI 0.68-0.89),MAE为1.9 dB(95%CI 1.6-2.4 dB);估算10-2 PSD的R2为0.69(95%CI 0.55-0.81),MAE为1.5 dB(95%CI 1.2-1.9 dB)。
  • 联合模型估算24-2 MD的R2为0.79(95%CI 0.72-0.84),MAE为2.1 dB(95%CI 1.8-2.5 dB);估算24-2 PSD的R2为0.68(95%CI 0.53-0.79),MAE为1.5 dB(95%CI 1.3-1.9 dB)。
  • 平均厚度线性回归估算10-2 MD的R2为0.61(95%CI 0.47-0.71),MAE为3.0 dB(95%CI 2.5-3.5 dB);估算10-2 PSD的R2为0.46(95%CI 0.31-0.60),MAE为2.3 dB(95%CI 1.8-2.7 dB)。
  • 平均厚度线性回归估算24-2 MD的R2为0.41(95%CI 0.26-0.57),MAE为3.4 dB(95%CI 2.7-4.5 dB);估算24-2 PSD的R2为0.38(95%CI 0.20-0.57),MAE为2.4 dB(95%CI 2.0-2.8 dB)。
  • GCIPL估算10-2 MD的R2为0.79,GCC估算24-2 MD的R2为0.75,分别为对应视野指标中表现最高的单层模型。

临床怎么用

  • 深度学习模型可基于黄斑OCT厚度图估算视野MD与PSD,辅助临床个体化安排视野检查。

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 deep learning (DL) systems estimating visual function from macula-centered spectral-domain (SD) OCT images. DESIGN: Evaluation of a diagnostic technology. PARTICIPANTS: A total of 2408 10-2 visual field (VF) SD OCT pairs and 2999 24-2 VF SD OCT pairs collected from 645 healthy and glaucoma subjects (1222 eyes). METHODS: Deep learning models were trained on thickness maps from Spectralis macula SD OCT to estimate 10-2 and 24-2 VF mean deviation (MD) and pattern standard deviation (PSD). Individual and combined DL models were trained using thickness data from 6 layers (retinal nerve fiber layer [RNFL], ganglion cell layer [GCL], inner plexiform layer [IPL], ganglion cell-IPL [GCIPL], ganglion cell complex [GCC] and retina). Linear regression of mean layer thicknesses were used for comparison. MAIN OUTCOME MEASURES: Deep learning models were evaluated using R2 and mean absolute error (MAE) compared with 10-2 and 24-2 VF measurements. RESULTS: Combined DL models estimating 10-2 achieved R2 of 0.82 (95% confidence interval [CI], 0.68-0.89) for MD and 0.69 (95% CI, 0.55-0.81) for PSD and MAEs of 1.9 dB (95% CI, 1.6-2.4 dB) for MD and 1.5 dB (95% CI, 1.2-1.9 dB) for PSD. This was significantly better than mean thickness estimates for 10-2 MD (0.61 [95% CI, 0.47-0.71] and 3.0 dB [95% CI, 2.5-3.5 dB]) and 10-2 PSD (0.46 [95% CI, 0.31-0.60] and 2.3 dB [95% CI, 1.8-2.7 dB]). Combined DL models estimating 24-2 achieved R2 of 0.79 (95% CI, 0.72-0.84) for MD and 0.68 (95% CI, 0.53-0.79) for PSD and MAEs of 2.1 dB (95% CI, 1.8-2.5 dB) for MD and 1.5 dB (95% CI, 1.3-1.9 dB) for PSD. This was significantly better than mean thickness estimates for 24-2 MD (0.41 [95% CI, 0.26-0.57] and 3.4 dB [95% CI, 2.7-4.5 dB]) and 24-2 PSD (0.38 [95% CI, 0.20-0.57] and 2.4 dB [95% CI, 2.0-2.8 dB]). The GCIPL (R2 = 0.79) and GCC (R2 = 0.75) had the highest performance estimating 10-2 and 24-2 MD, respectively. CONCLUSIONS: Deep learning models improved estimates of functional loss from SD OCT imaging. Accurate estimates can help clinicians to individualize VF testing to patients.

来源:Ophthalmology,查看原文。版权归原作者所有。

引用

  1. Christopher M, Bowd C, Proudfoot JA, et al. Deep Learning Estimation of 10-2 and 24-2 Visual Field Metrics Based on Thickness Maps from Macula OCT. Ophthalmology. 2021. PMID: 33901527. doi:10.1016/j.ophtha.2021.04.022

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。