Construction and validation of a prediction model for nonalcoholic fatty liver disease based on machine learning

CHENG Linjie; YUAN Qing; LI Wansong; LIU Ying; YANG Lei

doi:10.16462/j.cnki.zhjbkz.2025.06.009

Volume 29 Issue 6

Jun. 2025

Turn off MathJax

Article Contents

Article Navigation > CHINESE JOURNAL OF DISEASE CONTROL & PREVENTION > 2025 > 29(6): 682-687

CHENG Linjie, YUAN Qing, LI Wansong, LIU Ying, YANG Lei. Construction and validation of a prediction model for nonalcoholic fatty liver disease based on machine learning[J]. CHINESE JOURNAL OF DISEASE CONTROL & PREVENTION, 2025, 29(6): 682-687. doi: 10.16462/j.cnki.zhjbkz.2025.06.009

Citation:

CHENG Linjie, YUAN Qing, LI Wansong, LIU Ying, YANG Lei. Construction and validation of a prediction model for nonalcoholic fatty liver disease based on machine learning[J]. CHINESE JOURNAL OF DISEASE CONTROL & PREVENTION, 2025, 29(6): 682-687. doi: 10.16462/j.cnki.zhjbkz.2025.06.009

Citation:

PDF( 2341 KB)

Construction and validation of a prediction model for nonalcoholic fatty liver disease based on machine learning

doi: 10.16462/j.cnki.zhjbkz.2025.06.009

CHENG Linjie¹,
YUAN Qing¹,
LI Wansong²,
LIU Ying^{3
,
,},
YANG Lei^{4
,
,}

1.
School of Public Health, North China University of Science and Technology, Tangshan 063210, China
2.
The First Clinical Medical College, Hebei Medical University, Shijiazhuang 050000, China
3.
Department of Academic Affairs, the First Hospital of Hebei Medical University, Shijiazhuang 050023, China
4.
School of Public Health, Hebei Medical University, Shijiazhuang 050017, China

Funds:

A Cohort Study of Natural Population Health Trends in Hebei Province 226Z7705G

More Information

Corresponding author: LIU Ying, E-mail: wayymbb@126.com; YANG Lei, E-mail: yanglei1127@hebmu.edu.cn
Received Date: 2025-01-06
Rev Recd Date: 2025-04-12

Available Online: 2025-07-07

Publish Date: 2025-06-10

Abstract

Abstract

Objective This study aimed to construct and validate machine learning (ML) models for predicting nonalcoholic fatty liver disease (NAFLD), screen out the optimal model, and interpret it through the SHapley Additive exPlanations (SHAP) framework. Methods The data in the National Health and Nutrition Examination Surve database from January 2017 to March 2020 were randomly divided into a training set and a test set at a ratio of 7∶3. The least absolute shrinkage and selection operator regression was employed for feature selection, and six algorithms were used to construct the prediction models. The models were evaluated using the area under curve (AUC) and interpreted by the calibration curves, the decision curve analysis, variable importance plot, and SHAP plot. Results Of the 6 918 participants, 3 974 (57.44%) were diagnosed with NAFLD. The overall performance of eXtreme gradient boosting (XGBoost) model was better than other models, with an AUC of 0.851, an accuracy of 0.757, a sensitivity of 0.760 and a specificity of 0.754 on the test set. The main predictors were body roundness index, waist circumference, triglyceride glucose index, alanine aminotransferase, glycated hemoglobin and high-density lipoprotein cholesterol. In terms of model application, a user interface was developed for use by medical staff. Conclusions In this study, six ML models for predicting NAFLD were constructed and validated, among which XGBoost was more advantageous and could provide a reliable reference for early clinical screening of high-risk patients with NAFLD.
- Non-alcoholic fatty liver disease,
- Machine learning,
- Prediction model

FullText(HTML)

References(11)

References

[1]	Ji WD, Xue MY, Zhang YS, et al. A machine learning based framework to identify and classify non-alcoholic fatty liver disease in a large-scale population [J]. Front Public Health, 2022, 10: 846118. DOI: 10.3389/fpubh.2022.846118.
[2]	Perakakis N, Polyzos SA, Yazdani A, et al. Non-invasive diagnosis of non-alcoholic steatohepatitis and fibrosis with the use of omics and supervised learning: a proof of concept study [J]. Metabolism, 2019, 101: 154005. DOI: 10.1016/j.metabol.2019.154005.
[3]	Ma XF, Yang C, Liang K, et al. A predictive model for the diagnosis of non-alcoholic fatty liver disease based on an integrated machine learning method [J]. Am J Transl Res, 2021, 13(11): 12704-12713.
[4]	Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis [J]. J Clin Epidemiol, 1996, 49(12): 1373-1379. DOI: 10.1016/s0895-4356(96)00236-3.
[5]	Zhao YP, Li HL. Association of serum vitamin C with liver fibrosis in adults with nonalcoholic fatty liver disease [J]. Scand J Gastroenterol, 2022, 57(7): 872-877. DOI: 10.1080/00365521.2022.2041085.
[6]	Zou HX, Zhao FR, Lyu XH, et al. Development and validation of a new nomogram to screen for MAFLD [J]. Lipids Health Dis, 2022, 21(1): 133. DOI: 10.1186/s12944-022-01748-1.
[7]	Yuan KC, Tsai LW, Lee KH, et al. The development an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit [J]. Int J Med Inform, 2020, 141: 104176. DOI: 10.1016/j.ijmedinf.2020.104176.
[8]	Yi FL, Yang H, Chen DR, et al. XGBoost-SHAP-based interpretable diagnostic framework for Alzheimer′s disease [J]. BMC Med Inform Decis Mak, 2023, 23(1): 137. DOI: 10.1186/s12911-023-02238-9.
[9]	Dong BT, Zhang H, Duan YY, et al. Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma [J]. J Transl Med, 2024, 22(1): 455. DOI: 10.1186/s12967-024-05203-w.
[10]	Zuo D, Yang LX, Jin Y, et al. Machine learning-based models for the prediction of breast cancer recurrence risk [J]. BMC Med Inform Decis Mak, 2023, 23(1): 276. DOI: 10.1186/s12911-023-02377-z.
[11]	Ferraioli G, Soares Monteiro LB. Ultrasound-based techniques for the diagnosis of liver steatosis [J]. World J Gastroenterol, 2019, 25(40): 6053-6062. DOI: 10.3748/wjg.v25.i40.6053.