
浏览全部资源
扫码关注微信
陆军军医大学(第三军医大学)大坪医院消化内科,消化系统肿瘤精准防治重庆市重点实验室,重庆
文良志,E-mail:wenliangzhi@tmmu.edu.cn
陈东风,E-mail:chendf@tmmu.edu.cn
收稿:2025-10-15,
修回:2026-01-12,
录用:2026-01-14,
纸质出版:2026-02-15
移动端阅览
徐燕,侯文青,唐毓馨,等.利用可解释机器学习识别关键生物标志物预测代谢功能障碍相关脂肪性肝病患者的慢性肾病风险:基于NHANES数据库的研究[J].陆军军医大学学报,2026,48(03):354-365.DOI:10.16016/j.2097-0927.202510032.
XU Y,HOU W Q,TANG Y X,et al.Explainable machine learning for identifying key biomarkers in predicting chronic kidney disease risk among patients with metabolic dysfunction-associated fatty liver disease: A study based on NHANES database[J].J Army Med Univ,2026,48(03):354-365.DOI:10.16016/j.2097-0927.202510032.
徐燕,侯文青,唐毓馨,等.利用可解释机器学习识别关键生物标志物预测代谢功能障碍相关脂肪性肝病患者的慢性肾病风险:基于NHANES数据库的研究[J].陆军军医大学学报,2026,48(03):354-365.DOI:10.16016/j.2097-0927.202510032. DOI:
XU Y,HOU W Q,TANG Y X,et al.Explainable machine learning for identifying key biomarkers in predicting chronic kidney disease risk among patients with metabolic dysfunction-associated fatty liver disease: A study based on NHANES database[J].J Army Med Univ,2026,48(03):354-365.DOI:10.16016/j.2097-0927.202510032. DOI:
目的
2
基于机器学习算法构建代谢功能障碍相关脂肪性肝病(metabolic dysfunction-associated steatotic liver disease,MASLD)患者发生慢性肾病(chronic kidney disease, CKD)预测模型,绘制诊断列线图以指导临床诊治。
方法
2
本研究基于2007—2018年美国国家健康与营养调查(National Health and Nutrition Examination Survey, NHANES)数据库,筛选存在肝脏脂肪变性且至少伴有一项心脏代谢危险因素的MASLD患者作为分析样本。在排除合并慢性肾衰竭等患者后,最终纳入2 144例MASLD患者,收集其人口统计学、临床特征及实验室指标等34个变量。根据是否发生CKD,将患者分为CKD组(
n
=347)与非CKD组(
n
=1 797)。按7∶3比例将总样本分层抽样随机划分为训练集(
n
=1 501)和内部验证集(
n
=643),并纳入陆军军医大学大坪医院消化内科2024年1月至2025年10月期间符合既定纳入与排除标准的110例脂肪肝患者作为外部验证集。在模型构建方面,基于Lasso回归筛选变量后,分别纳入决策树(decision trees,DT)、极端梯度提升机(extreme gradient boosting, XGBoost)、K-最近邻算法(K-nearest neighbors,KNN)、朴素贝叶斯(Naive Bayes, NB)、支持向量机(support vector machine,SVM)、单隐藏层神经网络(neural networks, NNET)、梯度提升机(light gradient boosting machine,LightGBM)、随机森林(random forest,RF)共8种机器学习算法,以构建CKD风险预测模型。通过受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under curve,AUC)、敏感度、特异度等综合评价模型性能,并进一步通过校准曲线与临床决策曲线评估其临床应用价值。基于沙普利加性解释(Shapley additive explanation,SHAP)分析筛选出的重要生物标志物,构建了诊断列线图,并通过ROC曲线评估了其诊断准确性。
结果
2
本研究确定了10个生物标志物,包括收缩压、年龄、糖尿病史、血尿素氮、BMI、球蛋白、高密度胆固醇、中性粒细胞计数、血尿酸、γ-谷氨酰基转移酶,将上述生物标志物通过8种机器学习算法构建预测模型,综合评价发现LightGBM模型性能最好,AUC为0.871,敏感度为0.838,特异度为0.756,准确率为0.825,F1得分为0.889,Brier得分为0.091。校准曲线显示,基于LightGBM算法构建的预测模型在训练集和验证集一致性良好;临床决策曲线显示,使用该模型预测MASLD患者相关肾损害风险有助于临床决策,具有临床实用性。基于SHAP值构建诊断列线图,并以81.7分作为风险阈值,根据列线图风险阈值可将患者分为MASLD CKD高风险和低风险人群,并且列线图显示出良好的准确性和预测性能,其AUC为0.816。
结论
2
LightGBM模型较其他模型效能显著,其在临床早期预测MASLD患者CKD风险中具有应用潜力。
Objective
2
To construct a machine learning-based prediction model for chronic kidney disease(CKD) in patients with metabolic dysfunction-associated steatotic liver disease (MASLD) and to develop a diagnostic nomogram for guiding clinical management.
Methods
2
This study was based on the National Health and Nutrition Examination Survey (NHANES) database (2007 to 2018). Patients with MASLD
defined as hepatic steatosis with at least one cardiometabolic risk factor
were selected as the analysis sample. After excluding patients with chronic renal failure and other conditions
a total of 2 144 MASLD patients were included
and 34 variables (including demographics
clinical characteristics
and laboratory indicators) were collected. Patients were stratified into a chronic kidney disease (CKD) group (
n
=347) and a non-CKD group (
n
=1 797) based on CKD status. The total participants were randomly divided into a training set (
n
=1 501) and an internal validation set (
n
=643) using stratified sampling at a 7∶3 ratio. A prospective external validation cohort comprising 110 eligible fatty liver disease patients was recruited from our department during January 2024 and October 2025
adhering strictly to predefined inclusion and exclusion criteria Variables were screened using Least Absolute Shrinkage and Selection Operator (Lasso) regression. Eight machine learning algorithms were then employed to construct the CKD risk prediction models: decision trees (DT)
extreme gradient boosting (XGBoost)
K-nearest neighbors (KNN)
Naive Bayes (NB)
support vector machine (SVM)
single hidden layer neural networks (NNET)
light gradient boosting machine (LightGBM)
and random forest (RF). Model performance was comprehensively evaluated using receiver operating characteristic (ROC) curves area under the curve (AUC)
sensitivity
specificity
calibration curv
es
and clinical decision curves. Based on key biomarkers identified through SHAP analysis
a diagnostic nomogram was constructed
and its diagnostic accuracy was assessed using ROC curves.
Results
2
Using interpretable machine learning algorithms
we identified 10 key biomarkers: SBP
age
diabetes
BUN
BMI
globulin
HDL-C
neutrophil count
uric acid
and gamma-glutamyl transferase. These biomarkers were utilized to construct predictive models through 8 distinct machine learning algorithms. Comprehensive evaluation revealed that the LightGBM model achieved optimal performance
with an AUC of 0.871
a sensitivity of 0.838
a specificity of 0.756
an accuracy of 0.825
an F1 score of 0.889
and a Brier score of 0.091. Calibration curve analysis demonstrated good consistency of the LightGBM-based predictive model across both training and validation sets. Clinical decision curve analysis indicated significant clinical utility of this model in predicting kidney damage risk in patients with MASLD. Based on SHAP values
we developed a diagnostic nomogram with a risk cutoff score of 81.7. This nomogram enabled categorization of patients into high-risk and low-risk subsets for MASLD-related CKD
and exhibited favorable accuracy and robust predictive reliability with an AUC value of 0.816.
Conclusion
2
The LightGBM model significantly outperforms other models and demonstrates potential for application in early clinical prediction of CKD risk among MASLD patients.
RINELLA M E , LAZARUS J V , RATZIU V , et al . A multisociety Delphi consensus statement on new fatty liver disease nomenclature [J]. J Hepatol , 2023 , 79 ( 6 ): 1542 - 1556 . DOI: 10.1016/j.jhep.2023.06.003 http://dx.doi.org/10.1016/j.jhep.2023.06.003 .
YOUNOSSI Z M , KALLIGEROS M , HENRY L . Epidemiology of metabolic dysfunction-associated steatotic liver disease [J]. Clin Mol Hepatol , 2025 , 31 : S32 - S50 . DOI: 10.3350/cmh.2024.0431 http://dx.doi.org/10.3350/cmh.2024.0431 .
WANG Z L , XU M , HU Z G , et al . Prevalence of nonalcoholic fatty liver disease and its metabolic risk factors in women of different ages and body mass index [J]. Menopause , 2015 , 22 ( 6 ): 667 - 673 . DOI: 10.1097/GME.0000000000000352 http://dx.doi.org/10.1097/GME.0000000000000352 .
MIAO L , TARGHER G , BYRNE C D , et al . Current status and future trends of the global burden of MASLD [J]. Trends Endocrinol Metab , 2024 , 35 ( 8 ): 697 - 707 . DOI: 10.1016/j.tem.2024.02.007 http://dx.doi.org/10.1016/j.tem.2024.02.007 .
TARGHER G , TILG H , BYRNE C D . Non-alcoholic fatty liver disease: a multisystem disease requiring a multidisciplinary and holistic approach [J]. Lancet Gastroenterol Hepatol , 2021 , 6 ( 7 ): 578 - 588 . DOI: 10.1016/S2468-1253(21)00020-0 http://dx.doi.org/10.1016/S2468-1253(21)00020-0 .
SUN D Q , TARGHER G , BYRNE C D , et al . An international Delphi consensus statement on metabolic dysfunction-associated fatty liver disease and risk of chronic kidney disease [J]. Hepatobiliary Surg Nutr , 2023 , 12 ( 3 ): 386 - 403 . DOI: 10.21037/hbsn-22-421 http://dx.doi.org/10.21037/hbsn-22-421 .
JIANG F Y , WANG L J , YING H C , et al . Multisystem health comorbidity networks of metabolic dysfunction-associated steatotic liver disease [J]. Med , 2024 , 5 ( 11 ): 1413 - 1423.e3 . DOI: 10.1016/j.medj.2024.07.013 http://dx.doi.org/10.1016/j.medj.2024.07.013 .
TARGHER G , BYRNE C D , TILG H . MASLD: a systemic metabolic disorder with cardiovascular and malignant complications [J]. Gut , 2024 , 73 ( 4 ): 691 - 702 . DOI: 10.1136/gutjnl-2023-330595 http://dx.doi.org/10.1136/gutjnl-2023-330595 .
WEBSTER A C , NAGLER E V , MORTON R L , et al . Chronic kidney disease [J]. Lancet , 2017 , 389 ( 10075 ): 1238 - 1252 . DOI: 10.1016/S0140-6736(16)32064-5 http://dx.doi.org/10.1016/S0140-6736(16)32064-5 .
ROMAGNANI P , AGARWAL R , CHAN J C N , et al . Chronic kidney disease [J]. Nat Rev Dis Primers , 2025 , 11 : 8 . DOI: 10.1038/s41572-024-00589-9 http://dx.doi.org/10.1038/s41572-024-00589-9 .
ZHOU X D , ZHENG M H . Cardiovascular-kidney-metabolic syndrome and MASLD: integrating medical perspectives [J]. Nat Rev Cardiol , 2025 , 22 ( 11 ): 843 . DOI: 10.1038/s41569-025-01199-y http://dx.doi.org/10.1038/s41569-025-01199-y .
MANTOVANI A , ZAZA G , BYRNE C D , et al . Nonalcoholic fatty liver disease increases risk of incident chronic kidney disease: a systematic review and meta-analysis [J]. Metabolism , 2018 , 79 : 64 - 76 . DOI: 10.1016/j.metabol.2017.11.003 http://dx.doi.org/10.1016/j.metabol.2017.11.003 .
WEI S S , SONG J , XIE Y J , et al . Metabolic dysfunction-associated fatty liver disease can significantly increase the risk of chronic kidney disease in adults with type 2 diabetes [J]. Diabetes Res Clin Pract , 2023 , 197 : 110563 . DOI: 10.1016/j.diabres.2023.110563 http://dx.doi.org/10.1016/j.diabres.2023.110563 .
GREENER J G , KANDATHIL S M , MOFFAT L , et al . A guide to machine learning for biologists [J]. Nat Rev Mol Cell Biol , 2022 , 23 ( 1 ): 40 - 55 . DOI: 10.1038/s41580-021-00407-0 http://dx.doi.org/10.1038/s41580-021-00407-0 .
CHOWDHURY S U , SAYEED S , RASHID I , et al . Shapley-additive-explanations-based factor analysis for dengue severity prediction using machine learning [J]. J Imaging , 2022 , 8 ( 9 ): 229 . DOI: 10.3390/jimaging8090229 http://dx.doi.org/10.3390/jimaging8090229 .
CURTIN L R , MOHADJER L K , DOHRMANN S M , et al . National health and nutrition examination survey: sample design, 2007-2010 [J]. Vital Health Stat 2 , 2013 ( 160 ): 1 - 23 .
BEDOGNI G , BELLENTANI S , MIGLIOLI L , et al . The fatty liver index: a simple and accurate predictor of hepatic steatosis in the general population [J]. BMC Gastroenterol , 2006 , 6 ( 1 ): 33 . DOI: 10.1186/1471-230X-6-33 http://dx.doi.org/10.1186/1471-230X-6-33 .
CARLI F , SABATINI S , GAGGINI M , et al . Fatty liver index(FLI) identifies not only individuals with liver steatosis but also at high cardiometabolic risk [J]. Int J Mol Sci , 2023 , 24 ( 19 ): 14651 . DOI: 10.3390/ijms241914651 http://dx.doi.org/10.3390/ijms241914651 .
NOMURA T , ONO M , KOBAYASHI K , et al . Validation of fatty liver index as a predictor of hepatic steatosis in Asian populations: Impact of alcohol consumption and sex [J]. Hepatol Res , 2023 , 53 ( 10 ): 968 - 977 . DOI: 10.1111/hepr.13935 http://dx.doi.org/10.1111/hepr.13935 .
WANG Y Y , YU Y T , ZHANG H J , et al . Cardiovascular and renal burdens among patients with MAFLD and NAFLD in China [J]. Front Endocrinol , 2022 , 13 : 968766 . DOI: 10.3389/fendo.2022.968766 http://dx.doi.org/10.3389/fendo.2022.968766 .
NDUMELE C E , NEELAND I J , TUTTLE K R , et al . A synopsis of the evidence for the science and clinical management of cardiovascular-kidney-metabolic(CKM) syndrome: a scientific statement from the American Heart Association [J]. Circulation , 2023 , 148 ( 20 ): 1636 - 1664 . DOI: 10.1161/CIR.0000000000001186 http://dx.doi.org/10.1161/CIR.0000000000001186 .
THEOFILIS P , VORDONI A , KALAITZIDIS R G . Interplay between metabolic dysfunction-associated fatty liver disease and chronic kidney disease: Epidemiology, pathophysiologic mechanisms, and treatment considerations [J]. World J Gastroenterol , 2022 , 28 ( 39 ): 5691 - 5706 . DOI: 10.3748/wjg.v28.i39.5691 http://dx.doi.org/10.3748/wjg.v28.i39.5691 .
BYRNE C D , TARGHER G . NAFLD as a driver of chronic kidney disease [J]. J Hepatol , 2020 , 72 ( 4 ): 785 - 801 . DOI: 10.1016/j.jhep.2020.01.013 http://dx.doi.org/10.1016/j.jhep.2020.01.013 .
PARK J , KIM H J , KIM J , et al . Predictive value of serum albumin-to-globulin ratio for incident chronic kidney disease: a 12-year community-based prospective study [J]. PLoS One , 2020 , 15 ( 9 ): e0238421 . DOI: 10.1371/journal.pone.0238421 http://dx.doi.org/10.1371/journal.pone.0238421 .
JORCH S K , KUBES P . An emerging role for neutrophil extracellular traps in noninfectious disease [J]. Nat Med , 2017 , 23 ( 3 ): 279 - 287 . DOI: 10.1038/nm.4294 http://dx.doi.org/10.1038/nm.4294 .
KIOURTIS C , TERRADAS-TERRADAS M , GEE L M , et al . Hepatocellular senescence induces multi-organ senescence and dysfunction via TGFβ [J]. Nat Cell Biol , 2024 , 26 ( 12 ): 2075 - 2083 . DOI: 10.1038/s41556-024-01543-3 http://dx.doi.org/10.1038/s41556-024-01543-3 .
CHUNG G E , JEONG S M , CHO E J , et al . The association of fatty liver index and BARD score with all-cause and cause-specific mortality in patients with type 2 diabetes mellitus: a nationwide population-based study [J]. Cardiovasc Diabetol , 2022 , 21 ( 1 ): 273 . DOI: 10.1186/s12933-022-01691-6 http://dx.doi.org/10.1186/s12933-022-01691-6 .
0
浏览量
2
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621