Related Experiment Video
Updated: Jun 6, 2025
06:09
Author Spotlight: Advancing Hepatic Fibrosis Diagnosis Using Magnetic Resonance Elastography and AI
Published on: July 21, 2023
1.1K
Computed tomography-based radiomics and body composition model for predicting hepatic decompensation.
Yashbir Singh1, John E Eaton2, Sudhakar K Venkatesh1
1Department of Radiology, Mayo Clinic, Rochester, MN 55905, USA.
Oncotarget
|November 22, 2024
Summary
Computational radiomics effectively predicts hepatic decompensation in primary sclerosing cholangitis (PSC) patients using CT scans. This advanced imaging analysis shows high accuracy in identifying disease progression, offering a promising tool for patient management.
Area of Science:
- Medical Imaging
- Hepatology
- Artificial Intelligence in Medicine
Background:
- Primary sclerosing cholangitis (PSC) is a chronic liver disease causing bile duct inflammation and scarring, potentially leading to cirrhosis and hepatic decompensation.
- Early prediction of hepatic decompensation is crucial for managing PSC patients and improving outcomes.
Purpose of the Study:
- To evaluate the efficacy of computational radiomics in predicting hepatic decompensation in PSC patients.
- To assess the performance of a deep learning-based body composition model for this prediction task.
Main Methods:
- Utilized a deep learning model (body composition model) to quantify four compartments from CT scans: subcutaneous adipose tissue (SAT), skeletal muscle (SKM), visceral adipose tissue (VAT), and intermuscular adipose tissue (IMAT).
- Extracted radiomics features from these body composition compartments to develop a predictive model.
- Validated the model's performance on independent cohorts.
Main Results:
- The radiomics-based predictive model achieved high performance in validation cohorts.
- Achieved an accuracy score of 0.97, precision of 1.0, and an AUC score of 0.97 for predicting hepatic decompensation.
- Demonstrated the potential of computational radiomics in identifying patients at risk.
Conclusions:
- Computational radiomics using CT-derived body composition features shows significant promise for predicting hepatic decompensation in PSC.
- The developed model exhibits high accuracy, but further research is necessary to confirm clinical utility and address limitations in predicting future events.
