Application of Interpretable Machine Learning Models to Predict the Risk Factors of HBV-Related Liver Cirrhosis in
Wei Xia1,2, Yafeng Tan1, Bing Mei1
1Department of Laboratory Medicine, Jingzhou Hospital Affiliated to Yangtze University, Jingzhou, Hubei, People's Republic of China.
Insights
Machine learning models can predict hepatitis B virus-related liver cirrhosis (HBV-LC) risk in chronic hepatitis B (CHB) patients using routine data. The random forest model showed the best performance, aiding early diagnosis and clinical decisions.
Area of Science:
- Hepatology and Artificial Intelligence
- Clinical Informatics
- Predictive Modeling in Medicine
Background:
- Chronic hepatitis B (CHB) is a major global health concern, frequently progressing to hepatitis B virus-related liver cirrhosis (HBV-LC).
- Early identification of HBV-LC risk is crucial for timely and effective patient management.
- Existing prediction methods may lack accuracy or accessibility in resource-limited settings.
Purpose of the Study:
- To develop and compare nine machine learning (ML) models for predicting HBV-LC risk in CHB patients.
- To identify key clinical and laboratory variables predictive of HBV-LC development.
- To create an interpretable and accurate ML tool for early HBV-LC risk assessment.
Main Methods:
- Retrospective analysis of 777 CHB patients, with 50.45% progressing to HBV-LC.
- Utilized 52 clinical/laboratory variables, employing multiple imputation for missing data.
- Applied LASSO regression and Boruta algorithm for feature selection, identifying 24 key predictors.
- Evaluated nine ML models including Random Forest (RF), XGBoost, and Logistic Regression (LR).
- Assessed model performance using AUC, Brier score, accuracy, sensitivity, specificity, and F1 score, with cross-validation for tuning.
Main Results:
- The Random Forest (RF) model achieved superior performance with an AUC of 0.992 (training) and 0.907 (validation).
- A reconstructed model based on RF demonstrated strong predictive power with an AUC of 0.863 on the independent testing set.
- SHAP analysis identified RPR, PLT, HBV DNA, ALT, and TBA as critical predictors for HBV-LC.
- Calibration curves and decision curve analysis confirmed the RF model's accuracy and clinical utility.
Conclusions:
- An interpretable machine learning model, particularly Random Forest, can effectively predict HBV-LC risk in CHB patients using routine clinical data.
- The developed model enhances early risk identification, supporting clinical decision-making, especially in resource-limited environments.
- Key predictors like RPR, PLT, HBV DNA, ALT, and TBA offer valuable insights into HBV-LC pathogenesis and progression.
Abstract:
Chronic hepatitis B (CHB) infection represents a significant global public health issue, often leading to hepatitis B virus (HBV)-related liver cirrhosis (HBV-LC) with poor prognoses. Early identification of HBV-LC risk is essential for timely intervention. This study develops and compares nine machine learning (ML) models to predict HBV-LC risk in CHB patients using routine clinical and laboratory data. A retrospective analysis was conducted involving 777 CHB patients, with 50.45% (392/777) progressing to HBV-LC. Admission data consisted of 52 clinical and laboratory variables, with missing values addressed using multiple imputation. Feature selection utilized Least Absolute Shrinkage and Selection Operator (LASSO) regression and the Boruta algorithm, identifying 24 key variables. The evaluated ML models included XGBoost, logistic regression (LR), LightGBM, random forest (RF), AdaBoost, Gaussian naive Bayes (GNB), multilayer perceptron (MLP), support vector machine (SVM), and k-nearest neighbors (KNN). The data set was partitioned into an 80% training set (n = 621) and a 20% independent testing set (n = 156). Cross-validation (CV) facilitated hyperparameter tuning and internal validation of the optimal model. Performance metrics included the area under the receiver operating characteristic curve (AUC), Brier score, accuracy, sensitivity, specificity, and F1 score. The RF model demonstrated superior performance, with AUCs of 0.992 (training) and 0.907 (validation), while the reconstructed model achieved AUCs of 0.944 (training) and 0.945 (validation), maintaining an AUC of 0.863 in the testing set. Calibration curves confirmed a strong alignment between observed and predicted probabilities. Decision curve analysis indicated that the RF model provided the highest net benefit across threshold probabilities. The SHAP algorithm identified RPR, PLT, HBV DNA, ALT, and TBA as critical predictors. This interpretable ML model enhances early HBV-LC prediction and supports clinical decision-making in resource-limited settings.
More Related Videos
12:24A Three-Dimensional Spheroid Model to Investigate the Tumor-Stromal Interaction in Hepatocellular Carcinoma
Published on: September 30, 2021
11:34A Competent Hepatocyte Model Examining Hepatitis B Virus Entry through Sodium Taurocholate Cotransporting Polypeptide as a Therapeutic Target
Published on: May 10, 2022
Related Concept Videos
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
01:22Effect of Hepatic Disease on Pharmacokinetics: Pathophysiologic Assessment and Liver Function Test
01:25Hepatitis