文献速递炎症性肠病消化内镜

基于卷积神经网络的机器学习算法在内镜下溃疡性结肠炎严重程度预测中的诊断准确性:系统综述与Meta分析

Diagnostic accuracy of convolutional neural network-based machine learning algorithms in endoscopic severity prediction of ulcerative colitis: a systematic review and meta-analysis.
PubMed 文献 · Gastrointest Endosc · 2023-04-23

本Meta分析纳入12项研究,评估基于卷积神经网络(CNN)的机器学习算法在内镜图像上预测溃疡性结肠炎(UC)严重程度的汇总诊断准确性。结果显示汇总准确率91.5%、灵敏度82.8%、特异度92.4%、阳性预测值86.6%、阴性预测值88.6%。亚组分析显示使用UCEIS评分系统较MES评分系统在灵敏度和阳性预测值上显著更优。

要点

  • 基于CNN的机器学习算法在内镜下评估UC严重程度方面显示出较高的汇总诊断准确性。
  • 使用UCEIS评分训练CNN的灵敏度和阳性预测值显著优于MES评分系统。
  • 作者认为需进一步研究以在真实临床环境中确立这些发现。

核心结论

  • 基于CNN的机器学习算法在内镜下UC严重程度评估中表现出优异的汇总诊断准确性参数。
  • 在CNN训练中使用UCEIS评分可能比MES评分获得更好的结果。
  • 作者认为需进一步研究以在真实临床环境中确立这些发现。

研究设计与样本

  • 系统综述与Meta分析,检索MEDLINE、Scopus和Embase数据库(2022年6月),最终纳入12项研究。
  • 采用随机效应模型进行标准Meta分析,使用I²统计量评估异质性。

主要发现

  • CNN算法预测UC内镜严重程度的汇总准确率为91.5%(95%CI 88.3-93.8;I²=84%)。
  • 汇总灵敏度为82.8%(95%CI 78.3-86.5;I²=89%),汇总特异度为92.4%(95%CI 89.4-94.6;I²=84%)。
  • 汇总阳性预测值为86.6%(95%CI 82.3-90;I²=89%),汇总阴性预测值为88.6%(95%CI 85.7-91;I²=78%)。
  • 亚组分析显示UCEIS组灵敏度93.6%(95%CI 87.5-96.8;I²=77%)vs MES组82%(95%CI 75.6-87;I²=89%),P=0.003。
  • UCEIS组阳性预测值93.6%(95%CI 88.7-96.4;I²=68%)vs MES组83.6%(95%CI 76.8-88.8;I²=77%),P=0.007。

临床怎么用

  • 原文未据此给出临床推荐。

局限性

  • 作者指出需进一步研究以在真实临床环境中确立这些发现。
  • 纳入研究的汇总分析存在异质性:准确率I²=84%、灵敏度I²=89%、特异度I²=84%、阳性预测值I²=89%、阴性预测值I²=78%。

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 AND AIMS: Endoscopic assessment of ulcerative colitis (UC) can be performed by using the Mayo Endoscopic Score (MES) or the Ulcerative Colitis Endoscopic Index of Severity (UCEIS). In this meta-analysis, we assessed the pooled diagnostic accuracy parameters of deep machine learning by means of convolutional neural network (CNN) algorithms in predicting UC severity on endoscopic images. METHODS: Databases including MEDLINE, Scopus, and Embase were searched in June 2022. Outcomes of interest were the pooled accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Standard meta-analysis methods used the random-effects model, and heterogeneity was assessed using the I2statistics. RESULTS: Twelve studies were included in the final analysis. The pooled diagnostic parameters of CNN-based machine learning algorithms in endoscopic severity assessment of UC were as follows: accuracy 91.5% (95% confidence interval [CI], 88.3-93.8; I2 = 84%), sensitivity 82.8% (95% CI, 78.3-86.5; I2 = 89%), specificity 92.4% (95% CI, 89.4-94.6; I2 = 84%), PPV 86.6% (95% CI, 82.3-90; I2 = 89%), and NPV 88.6% (95% CI, 85.7-91; I2 = 78%). Subgroup analysis revealed significantly better sensitivity and PPV with the UCEIS scoring system compared with the MES (93.6% [95% CI, 87.5-96.8; I2 = 77%] vs 82% [95% CI, 75.6-87; I2 = 89%], P = .003, and 93.6% [95% CI, 88.7-96.4; I2 = 68%] vs 83.6% [95% CI, 76.8-88.8; I2 = 77%], P = .007, respectively). CONCLUSIONS: CNN-based machine learning algorithms demonstrated excellent pooled diagnostic accuracy parameters in the endoscopic severity assessment of UC. Using UCEIS scores in CNN training might offer better results than the MES. Further studies are warranted to establish these findings in real clinical settings.

来源:Gastrointest Endosc,查看原文。版权归原作者所有。

引用

  1. Jahagirdar V, Bapaye J, Chandan S, et al. Diagnostic accuracy of convolutional neural network-based machine learning algorithms in endoscopic severity prediction of ulcerative colitis: a systematic review and meta-analysis. Gastrointest Endosc. 2023. PMID: 37094691. doi:10.1016/j.gie.2023.04.2074

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。