文献速递肝病

机器学习模型预测肝硬化失代偿风险:一项验证研究

Predicting the Risk of Decompensation in Patients With Cirrhosis: A Validated Machine Learning Approach.
PubMed 文献 · Am J Gastroenterol · 2026-03-25

该研究利用575例代偿期肝硬化患者数据开发机器学习模型RODIC,基于白蛋白、胆红素和Fib-4预测3年内失代偿风险,训练集AUC为0.86,并在超过2100例国际独立人群中验证,AUC为0.67至0.80。

要点

  • RODIC模型仅用白蛋白、胆红素和Fib-4即可预测3年内失代偿风险。
  • 训练集AUC 0.86,验证集AUC 0.67-0.80,校准良好。
  • 模型效果不受肝硬化病因影响,丙肝患者无论是否获持续病毒学应答均可靠。

核心结论

  • 基于常规临床和实验室数据的机器学习模型RODIC能准确量化代偿期肝硬化患者的失代偿风险。

研究设计与样本

  • 使用诺丁汉(英国)和摩德纳(意大利)训练队列575例代偿期肝硬化患者的常规临床和实验室数据,中位随访4年,采用随机森林分类器开发模型。
  • 在都柏林(爱尔兰)、梅努菲亚(埃及)、利兹(英国)和大垣(日本)的独立国际人群(超过2100例患者)中验证模型。

主要发现

  • 训练队列中22%的患者发生肝失代偿。
  • RODIC模型基于白蛋白、胆红素和Fib-4值,预测3年内失代偿风险,训练集AUC为0.86,加权F1分数为0.82。
  • 验证集中RODIC的AUC范围为0.67至0.80,加权F1分数范围为0.70至0.81。
  • 模型效果不受肝硬化病因影响;对于丙型肝炎病毒阳性患者,无论是否获得持续病毒学应答,RODIC均保持可靠。

关键数值与推荐

  • RODIC模型使用白蛋白、胆红素和Fib-4值计算3年内失代偿风险,在线工具免费访问:https://antonkaly.pythonanywhere.com/predict。
  • 训练集AUC=0.86,加权F1分数=0.82;验证集AUC 0.67-0.80,加权F1分数0.70-0.81。

临床怎么用

  • 临床可使用RODIC在线工具,基于常规实验室指标评估代偿期肝硬化患者3年内失代偿风险,辅助分层管理。

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

INTRODUCTION: There is an unmet need for simple tools to predict development of hepatic decompensation among patients with compensated cirrhosis. We applied machine learning to several international Data sets to develop and validate a straightforward predictive model of decompensation. METHODS: We used routinely available clinical and laboratory data from 575 patients with compensated cirrhosis from the training cohorts in Nottingham (United Kingdom) and Modena (Italy), with a median follow-up of 4 years. Based on these data, we developed a predictive model using a random forest classifier and validated it across independent international populations involving over 2,100 patients from Dublin (Ireland), Menoufia (Egypt), Leeds (UK), and Ogaki (Japan). RESULTS: In the training cohorts, 22% of patients developed liver decompensation. Using machine learning, we developed risk of decompensation in cirrhosis (RODIC), a well-calibrated model based on albumin, bilirubin, and the Fib-4 value (area under the receiver operating characteristic curve = 0.86; weighted F1 score = 0.82) which predicts the risk of decompensation within 3 years (access free of charge at: https://antonkaly.pythonanywhere.com/predict ). RODIC showed strong performance across all validation sets, with area under the receiver operating characteristic curve scores ranging from 0.67 to 0.80 and weighted F1 scores from 0.70 to 0.81. Moreover, the model was effective regardless of cirrhosis etiology. For hepatitis C virus-positive patients, RODIC remained reliable irrespective of whether they achieved sustained virologic response. DISCUSSION: Our validated machine learning model based on readily available clinical, and laboratory features accurately quantitates the risk of liver decompensation in patients with compensated hepatic cirrhosis.

来源:Am J Gastroenterol,查看原文。版权归原作者所有。

引用

  1. Johnson PJ, Kalyuzhnyy A, Grauslys A, et al. Predicting the Risk of Decompensation in Patients With Cirrhosis: A Validated Machine Learning Approach. Am J Gastroenterol. 2026. PMID: 41879220. doi:10.14309/ajg.0000000000004004

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。