Related Experiment Video
Updated: Apr 8, 2026
Mouse Model of Metabolic Dysfunction-Associated Steatotic Liver Disease with Fibrosis
Published on: July 18, 2025
Development and Validation of a Machine Learning-Based Prediction Model for Cardiovascular Disease in Patients with
1Department of Infectious Diseases, Jiashan County First People's Hospital, Jiashan, China.
Insights
Metabolic dysfunction-associated fatty liver disease (MAFLD) patients have high cardiovascular disease (CVD) risk. Liver fibrosis (NFS) and albumin levels are better CVD predictors than traditional factors, enabling early intervention.
Area of Science:
- Cardiology
- Hepatology
- Machine Learning
- Public Health
Background:
- Metabolic dysfunction-associated fatty liver disease (MAFLD) significantly elevates cardiovascular disease (CVD) risk.
- Traditional CVD risk assessment tools may underestimate risks in MAFLD patients due to complex liver-heart interactions.
- There is a need for improved CVD risk prediction models tailored for MAFLD.
Purpose of the Study:
- To develop and validate machine learning models for predicting CVD risk specifically in MAFLD patients.
- To identify key predictors of CVD in MAFLD, focusing on liver-specific markers.
- To compare the predictive power of liver-specific markers versus traditional cardiovascular risk factors.
Main Methods:
- Analysis of cross-sectional NHANES 2017-2023 data from 6828 MAFLD adults.
- Random split of data into 70% training and 30% validation sets.
- Development of machine learning models using Least Absolute Shrinkage and Selection Operator regression, with SHapley Additive exPlanations for feature importance.
Main Results:
- Liver-specific markers, particularly the nonalcoholic fatty liver disease Fibrosis Score (NFS) and albumin, were superior predictors of CVD risk compared to traditional factors like hypertension, diabetes, and smoking.
- Age, NFS (especially > -1.0), and low albumin (<3.5 g/dL) were identified as the strongest drivers of CVD risk in MAFLD patients.
- The developed model demonstrated significant predictive capacity, with NFS showing clear threshold effects.
Conclusions:
- Traditional CVD risk assessment is insufficient for MAFLD patients.
- Hepatic fibrosis (NFS) and liver synthetic function (albumin) are critical for accurate CVD risk stratification in MAFLD.
- Integrated liver-heart assessment and routine hepatic fibrosis evaluation are recommended for MAFLD patients to enable early, targeted preventive interventions.
Background/Aim:
Metabolic dysfunction-associated fatty liver disease (MAFLD) is strongly associated with increased cardiovascular disease (CVD) risk. However, traditional cardiovascular risk assessment tools may inadequately capture the complex pathophysiology linking hepatic and CVD in MAFLD patients. This study aimed to develop and validate machine learning models to predict CVD risk in MAFLD patients.
Materials And Methods:
This cross-sectional study analyzed NHANES 2017-2023 data, with participants randomly split into training (70%) and validation (30%) sets.
Results:
A total of 6828 adults with MAFLD were included (13.1% with prevalent CVD). Least Absolute Shrinkage and Selection Operator regression identified 14 key predictors, with age, nonalcoholic fatty liver disease Fibrosis Score (NFS), and albumin emerging as the most influential. Critically, liver-specific markers (NFS: mean |SHAP| = 0.0299; albumin: 0.0289) demonstrated superior predictive capacity compared to traditional cardiovascular risk factors (hypertension: 0.0145; diabetes: 0.0107; smoking: 0.0035). SHapley Additive exPlanations analysis revealed that older age, higher NFS (particularly > -1.0), and lower albumin (<3.5 g/dL) were the strongest drivers of CVD risk, with NFS showing clear threshold effects.
Conclusion:
The findings confirm that traditional cardiovascular risk assessment approaches are insufficient for MAFLD patients, as liver-specific markers-particularly hepatic fibrosis (NFS) and liver synthetic function (albumin)-dominated cardiovascular risk prediction over conventional risk factors (hypertension, diabetes, smoking). This paradigm shift underscores the necessity of integrated liver-heart assessment in MAFLD and supports routine hepatic fibrosis evaluation for cardiovascular risk stratification. The model demonstrates immediate clinical applicability through existing electronic health records, enabling early identification of high-risk patients for targeted preventive interventions.
