Related Experiment Video
Updated: Oct 5, 2025
08:41
Novel In Vivo Micro-Computed Tomography Imaging Techniques for Assessing the Progression of Non-Alcoholic Fatty Liver Disease
Published on: March 24, 2023
1.3K
Computed Tomography-Based Radiomic Analysis for Preoperatively Predicting the Macrovesicular Steatosis Grade in
Shengnan Ding1, Weimin Yang2, Xiaodong Sun3
1Department of Radiology, The First Hospital of Jilin University, Changchun, Jilin 130021, China.
Biomed Research International
|February 1, 2022
Summary
Computed tomography (CT)-based radiomics effectively distinguishes between low and high grades of macrovesicular steatosis (MaS) in liver donors. This advanced radiomic model outperforms clinical factors, aiding in donor selection for liver transplantation.
Area of Science:
- Radiology
- Medical Imaging
- Oncology
Background:
- Macrovesicular steatosis (MaS) is a critical factor in liver transplant outcomes.
- Accurate grading of MaS is essential for selecting viable donor livers.
- Non-contrast computed tomography (CT) is routinely used in donor liver assessment.
Purpose of the Study:
- To evaluate the efficacy of CT-based radiomic analysis in differentiating low-grade (0/1) from high-grade (2/3) MaS.
- To compare the performance of a radiomic model against a clinical model for MaS grading.
Main Methods:
- Retrospective analysis of preoperative noncontrast CT scans from 150 liver transplant donors with biopsy-confirmed MaS.
- Segmentation of liver regions of interest (ROIs) and extraction of 402 radiomic features.
- Development of a radiomic model using multivariate logistic regression and comparison with a clinical model based on liver function tests.
Main Results:
- The radiomic model demonstrated strong discrimination for MaS grading, achieving an AUC of 0.907 in the training cohort and 0.906 in the testing cohort.
- The model showed good calibration, with predicted probabilities aligning well with observed outcomes.
- The radiomic model significantly outperformed the clinical model (AUC 0.784) in predicting MaS grade.
Conclusions:
- CT-based radiomic analysis is a highly effective, non-invasive tool for grading macrovesicular steatosis in potential liver donors.
- This radiomic approach offers a valuable improvement over conventional clinical assessments for donor liver evaluation.
- The findings support the integration of radiomics into standard protocols for assessing donor liver suitability in transplantation.
