Predicting Metabolic Dysfunction-Associated Steatotic Liver Disease using Machine Learning Methods

2025-10-25

AI Summary

MASLD affects approximately 33% of U.S. adults, and early identification is critical to enable effective lifestyle interventions. To address this, we propose MASER—a clinically interpretable and fairness-aware predictive model for MASLD risk stratification. MASER integrates sparse LASSO logistic regression with an equal opportunity post-processing technique, leveraging large-scale real-world electronic health records. It preserves clinical interpretability while substantially mitigating racial disparities in prediction performance. Compared to baseline models—including random forests, XGBoost, and neural networks—MASER enhances transparency via SHAP-based feature attribution and achieves an AUROC of 0.836, accuracy of 77.6%, and specificity of 94% after fairness optimization. To our knowledge, this is the first work to jointly employ sparse logistic regression and rigorous fairness constraints for MASLD prediction, achieving a substantive trade-off among predictive performance, algorithmic fairness, and clinical interpretability.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Human-in-the-loop Machine Learning
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Background: Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) affects ~33% of U.S. adults and is the most common chronic liver disease. Although often asymptomatic, progression can lead to cirrhosis. Early detection is important, as lifestyle interventions can prevent disease progression. We developed a fair, rigorous, and reproducible MASLD prediction model and compared it to prior methods using a large electronic health record database. Methods: We evaluated LASSO logistic regression, random forest, XGBoost, and a neural network for MASLD prediction using clinical feature subsets, including the top 10 SHAP-ranked features. To reduce disparities in true positive rates across racial and ethnic subgroups, we applied an equal opportunity postprocessing method. Results: This study included 59,492 patients in the training data, 24,198 in the validating data, and 25,188 in the testing data. The LASSO logistic regression model with the top 10 features was selected for its interpretability and comparable performance. Before fairness adjustment, the model achieved AUROC of 0.84, accuracy of 78%, sensitivity of 72%, specificity of 79%, and F1-score of 0.617. After equal opportunity postprocessing, accuracy modestly increased to 81% and specificity to 94%, while sensitivity decreased to 41% and F1-score to 0.515, reflecting the fairness trade-off. Conclusions: We developed the MASER prediction model (MASLD Static EHR Risk Prediction), a LASSO logistic regression model which achieved competitive performance for MASLD prediction (AUROC 0.836, accuracy 77.6%), comparable to previously reported ensemble and tree-based models. Overall, this approach demonstrates that interpretable models can achieve a balance of predictive performance and fairness in diverse patient populations.

Problem

• Predicting Metabolic Dysfunction-Associated Steatotic Liver Disease using machine learning
• Developing fair and interpretable models to reduce healthcare disparities
• Comparing multiple ML methods for early MASLD detection in EHR data

Innovation

• Used LASSO logistic regression with top features
• Applied equal opportunity postprocessing for fairness
• Compared multiple machine learning methods for prediction