文献速递青光眼神经眼科

深度学习分类器在SD-OCT上鉴别青光眼性与压迫性视神经病变

Discriminating glaucomatous and compressive optic neuropathy on spectral-domain optical coherence tomography with deep learning classifier.
PubMed 文献 · Br J Ophthalmol · 2020-02-25

本研究评估深度学习分类器基于GCIPL和RNFL的SD-OCT图像鉴别青光眼性视神经病变(GON)与压迫性视神经病变(CON)的性能。研究纳入80眼GON和54眼CON的SD-OCT图像集,提取瓶颈特征作为预测因子,五折交叉验证显示AUC为0.990(95%CI 0.982-0.999),敏感度97.9%,特异度92.6%,显著优于传统诊断参数。

要点

  • 深度学习分类器鉴别GON与CON的AUC达0.990,敏感度97.9%,特异度92.6%。
  • 分类器性能显著优于颞侧中缝征(AUC 0.804)、鼻上GCIPL厚度(AUC 0.815)和上方GCIPL厚度(AUC 0.776)。
  • 基于GCIPL和RNFL厚度图及偏差图的瓶颈特征可有效区分GON与CON。

核心结论

  • 深度学习分类器在SD-OCT上鉴别GON与CON的性能优于传统诊断参数。

研究设计与样本

  • 研究纳入80眼GON和54眼CON的SD-OCT图像集,提取GCIPL厚度图、GCIPL偏差图、RNFL厚度图和RNFL偏差图的瓶颈特征作为预测因子。
  • 采用五折交叉验证测试,计算AUC以验证诊断性能,并与传统诊断参数比较。

主要发现

  • 深度学习系统在五折交叉验证中AUC为0.990(95%CI 0.982至0.999),敏感度97.9%,特异度92.6%。
  • 颞侧中缝征AUC为0.804(95%CI 0.737至0.872),鼻上GCIPL厚度AUC为0.815(95%CI 0.734至0.896),上方GCIPL厚度AUC为0.776(95%CI 0.691至0.860)。
  • 深度学习分类器AUC显著大于其他参数(所有p<0.001)。

关键数值与推荐

  • 深度学习分类器AUC 0.990(95%CI 0.982至0.999),敏感度97.9%,特异度92.6%。
  • 颞侧中缝征AUC 0.804(95%CI 0.737至0.872)。
  • 鼻上GCIPL厚度AUC 0.815(95%CI 0.734至0.896)。
  • 上方GCIPL厚度AUC 0.776(95%CI 0.691至0.860)。

临床怎么用

  • 基于SD-OCT的深度学习分类器可辅助鉴别GON与CON,性能优于传统参数。

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 a deep learning classifier for differentiation of glaucomatous optic neuropathy (GON) from compressive optic neuropathy (CON) based on ganglion cell-inner plexiform layer (GCIPL) and retinal nerve fibre layer (RNFL) spectral-domain optical coherence tomography (SD-OCT). METHODS: Eighty SD-OCT image sets from 80 eyes of 80 patients with GON along with 81 SD-OCT image sets from 54 eyes of 54 patients with CON were compiled for the study. The bottleneck features extracted from the GCIPL thickness map, GCIPL deviation map, RNFL thickness map and RNFL deviation map were used as predictors for the deep learning classifier. The area under the receiver operating characteristic curve (AUC) was calculated to validate the diagnostic performance. The AUC with the deep learning classifier was compared with those for conventional diagnostic parameters including temporal raphe sign, SD-OCT thickness profile and standard automated perimetry. RESULTS: The deep learning system achieved an AUC of 0.990 (95% CI 0.982 to 0.999) with a sensitivity of 97.9% and a specificity of 92.6% in a fivefold cross-validation testing, which was significantly larger than the AUCs with the other parameters: 0.804 (95% CI 0.737 to 0.872) with temporal raphe sign, 0.815 (95% CI 0.734 to 0.896) with superonasal GCIPL and 0.776 (95% CI 0.691 to 0.860) with superior GCIPL thicknesses (all p<0.001). CONCLUSION: The deep learning classifier can outperform the conventional diagnostic parameters for discrimination of GON and CON on SD-OCT.

来源:Br J Ophthalmol,查看原文。版权归原作者所有。

引用

  1. Lee J, Kim JS, Lee HJ, et al. Discriminating glaucomatous and compressive optic neuropathy on spectral-domain optical coherence tomography with deep learning classifier. Br J Ophthalmol. 2020. PMID: 32098860. doi:10.1136/bjophthalmol-2019-314330

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。