Predicting Progression of Metabolic Dysfunction-associated Steatotic Liver Disease

This work focuses on the progression from metabolic dysfunction-associated fatty liver to metabolic dysfunction-associated steatohepatitis, a more serious prognosis that can lead to liver failure and death. Additional adverse progressed outcomes include hepatic failure, fibrosis, cirrhosis, and malignant neoplasm of liver and intrahepatic bile ducts. We explore the possibility of using different machine learning techniques, including logistic regression, XGBoost, random forest, and decision trees to predict the likelihood of progression. We use data from Massachusetts General Brigham to train our models, incorporating demographics, physical measurements, lab results, and doctor notes. As a result of this project, we our best model was an XGBoost classifier with an AUROC of 0.800 with random forest at a similar performance of 0.786. However, all of our models had low AUPRC and sensitivity, indicating both overfitting and an imbalanced dataset.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2025

Author dc:creator

• Li, Jonathan

Advisor dc:contributor.advisor

• Szolovits, Peter

Statement dc:rights

• In Copyright - Educational Use Permitted
• Copyright retained by author(s)

Licence dc:rights.uri

• https://rightsstatements.org/page/InC-EDU/1.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/162692

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/162692