文献速递糖尿病肥胖与代谢综合征

基于机器学习的生物标志物识别用于代谢功能障碍相关脂肪性肝病的早期诊断

Machine Learning-Based Biomarker Identification for Early Diagnosis of Metabolic Dysfunction-Associated Steatotic Liver Disease.
PubMed 文献 · J Clin Endocrinol Metab · 2025-01-01

随机森林模型预测MASLD准确率79.59%(P<.001)、特异度84.65%,预测肝纤维化准确率86.07%(P<.001)、灵敏度98.01%。

要点

  • 随机森林模型预测MASLD准确率79.59%(P<.001),特异度84.65%。
  • 随机森林模型预测肝纤维化准确率86.07%(P<.001),灵敏度98.01%。
  • 年龄、BMI和胰岛素在脂肪变性与硬度两个模型中均具统计学显著性。

核心结论

  • 评估涵盖人口学、代谢、血脂和标准生化类别的多种生物标志物,可能为诊断MASLD和肝纤维化提供有价值的初步线索。

研究设计与样本

  • 采用随机森林机器学习确定关键生物标志物(年龄、性别、种族、BMI、HbA1c、空腹血糖、胰岛素、总胆固醇、LDL胆固醇、HDL胆固醇、甘油三酯、ALT、AST、ALP、白蛋白、GGT、LDH、铁、总胆红素、总蛋白、尿酸、BUN和hs-CRP)纳入逻辑回归模型,预测脂肪变性(MASLD定义为受控衰减参数≥238 dB/m)和硬度(肝纤维化定义为肝脏硬度中位数≥7 kPa)。
  • 使用XGBoost和递归特征消除进行敏感性分析。

主要发现

  • 随机森林模型(最准确)预测MASLD的准确率为79.59%(P<.001),特异度为84.65%。
  • 随机森林模型预测肝纤维化的准确率为86.07%(P<.001),灵敏度为98.01%。
  • 脂肪变性和硬度模型均识别出统计学显著的生物标志物,其中年龄、BMI和胰岛素在两个模型中均显著。

关键数值与推荐

  • MASLD由受控衰减参数评分≥238 dB/m指示。
  • 肝纤维化由肝脏硬度中位数≥7 kPa指示。
  • 随机森林模型预测MASLD准确率79.59%(P<.001)、特异度84.65%;预测肝纤维化准确率86.07%(P<.001)、灵敏度98.01%。

临床怎么用

  • 原文未据此给出推荐。

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

CONTEXT: Metabolic dysfunction-associated steatotic liver disease (MASLD) is an umbrella term for simple hepatic steatosis and the more severe metabolic dysfunction-associated steatohepatitis. The current reliance on liver biopsy for diagnosis and a lack of validated biomarkers are major factors contributing to the overall burden of MASLD. OBJECTIVE: This study investigates the association between biomarkers and hepatic steatosis and stiffness measurements, measured by FibroScan®. METHODS: Data from the National Health and Nutritional Examination Survey (2017-2020) were collected for 15 560 patients. Propensity score matching balanced the data with a 1:1 case to control for age and sex allowing for preliminary trend assessment. Random Forest machine learning determined variable importance for the incorporation of key biomarkers (age, sex, race, BMI, HbA1c, plasma fasting glucose, insulin, total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, ALT, AST, ALP, albumin, GGT, LDH, iron, total bilirubin, total protein, uric acid, BUN, and hs-CRP) into logistic regression models predicting steatosis (MASLD indicated by a controlled attenuation parameter score of ≥238 dB/m) and stiffness (hepatic fibrosis indicated by a median liver stiffness ≥7 kPa). Sensitivity analysis using XGBoost and Recursive Feature Elimination was performed. RESULTS: The Random Forest models (the most accurate) predicted MASLD with 79.59% accuracy (P < .001) and specificity of 84.65% and predicted hepatic fibrosis with 86.07% accuracy (P < .001) and sensitivity of 98.01%. Both the steatosis and stiffness models identified statistically significant biomarkers, with age, BMI, and insulin appearing significant to both. CONCLUSION: These findings indicate that assessing a variety of biomarkers, across demographic, metabolic, lipid, and standard biochemistry categories, may provide valuable initial insights for diagnosing patients for MASLD and hepatic fibrosis.

来源:J Clin Endocrinol Metab,查看原文。版权归原作者所有。

引用

  1. Boullion J, Husein A, Agrawal A, et al. Machine Learning-Based Biomarker Identification for Early Diagnosis of Metabolic Dysfunction-Associated Steatotic Liver Disease. J Clin Endocrinol Metab. 2025. PMID: 39980343. doi:10.1210/clinem/dgaf111

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。