文献速递肝病

机器学习模型在识别非酒精性脂肪性肝病临床显著分期及非酒精性脂肪性肝病相关肝硬化方面优于无创检测

Machine learning models are superior to noninvasive tests in identifying clinically significant stages of NAFLD and NAFLD-related cirrhosis.
PubMed 文献 · Hepatology · 2022-08-09

研究纳入1370例接受肝活检、FibroScan和实验室检查的非酒精性脂肪性肝病患者,比较机器学习模型(逻辑回归、随机森林、人工神经网络)与FibroScan、FIB-4、FAST、NFS在预测肝纤维化分期(≥F2、≥F3、F4)及NASH伴显著纤维化(NAS≥4+≥F2)中的性能。结果显示所有机器学习模型尤其是随机森林的准确性和AUC主要高于传统无创检测,随机森林对≥F2、≥F3、F4的AUC分别为0.86、0.89、0.89。

要点

  • 随机森林对≥F2、≥F3、F4的AUC分别为0.86、0.89、0.89,高于FibroScan和FIB-4。
  • 随机森林对NASH+NAS≥4+≥F2的AUC为0.80,高于FAST、FIB-4和NFS。
  • 机器学习模型在敏感性、特异性、阳性预测值和阴性预测值方面整体优于传统无创检测。

核心结论

  • 机器学习模型整体性能优于FibroScan、FIB-4、FAST和NFS,可能成为识别非酒精性脂肪性肝病患者临床显著肝纤维化和肝硬化的有效工具。

研究设计与样本

  • 研究纳入1370例在美国多个中心接受肝活检、FibroScan和实验室检查的非酒精性脂肪性肝病患者,使用17个人口学/临床特征构建逻辑回归、随机森林和人工神经网络模型,80%队列用于训练、20%用于测试。

主要发现

  • 对于≥F2、≥F3、F4和NASH+NAS≥4+≥F2,所有机器学习模型尤其是随机森林的准确性和AUC主要高于FibroScan、FIB-4、FAST和NFS。
  • 随机森林 vs FibroScan和FIB-4对≥F2、≥F3、F4的AUC分别为(0.86 vs 0.81、0.78)、(0.89 vs 0.83、0.82)、(0.89 vs 0.86、0.85)。
  • 随机森林 vs FAST、FIB-4和NFS对NASH+NAS≥4+≥F2的AUC分别为(0.80 vs 0.77、0.66、0.63)。
  • 对于NASH+NAS≥4+≥F2,所有机器学习模型落在不确定区的百分比低于或相似于FIB-4和NFS。
  • 整体上,机器学习模型在敏感性、特异性、阳性预测值和阴性预测值方面优于传统无创检测。

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 AND AIMS: We assessed the performance of machine learning (ML) models in identifying clinically significant NAFLD-associated liver fibrosis and cirrhosis. APPROACH AND RESULTS: We implemented ML models including logistic regression (LR), random forest (RF), and artificial neural network to predict histological stages of fibrosis using 17 demographic/clinical features in 1370 patients with NAFLD who underwent liver biopsy, FibroScan, and labs within a 6-month period at multiple U.S. centers. Histological stages of fibrosis (≥F2, ≥F3, and F4) were predicted using ML, FibroScan liver stiffness measurements, and Fibrosis-4 index (FIB-4). NASH with significant fibrosis (NAS ≥ 4 + ≥F2) was assessed using ML, FibroScan-AST (FAST) score, FIB-4, and NAFLD fibrosis score (NFS). We used 80% of the cohort to train and 20% to test the ML models. For ≥F2, ≥F3, F4, and NASH + NAS ≥ 4 + ≥F2, all ML models, especially RF, had primarily higher accuracy and AUC compared with FibroScan, FIB-4, FAST, and NFS. AUC for RF versus FibroScan and FIB-4 for ≥F2, ≥F3, and F4 were (0.86 vs. 0.81, 0.78), (0.89 vs. 0.83, 0.82), and (0.89 vs. 0.86, 0.85), respectively. AUC for RF versus FAST, FIB-4, and NFS for NASH + NAS ≥ 4 + ≥F2 were (0.80 vs. 0.77, 0.66, 0.63). For NASH + NAS ≥ 4 + ≥F2, all ML models had lower/similar percentages within the indeterminate zone compared with FIB-4 and NFS. Overall, ML models performed better in sensitivity, specificity, positive predictive value, and negative predictive value compared with traditional noninvasive tests. CONCLUSIONS: ML models performed better overall than FibroScan, FIB-4, FAST, and NFS. ML could be an effective tool for identifying clinically significant liver fibrosis and cirrhosis in patients with NAFLD.

来源:Hepatology,查看原文。版权归原作者所有。

引用

  1. Chang D, Truong E, Mena EA, et al. Machine learning models are superior to noninvasive tests in identifying clinically significant stages of NAFLD and NAFLD-related cirrhosis. Hepatology. 2022. PMID: 35809234. doi:10.1002/hep.32655

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。