Related Experiment Video
Updated: May 30, 2025
13:35
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
Published on: March 21, 2021
10.3K
CompositIA: an open-source automated quantification tool for body composition scores from thoraco-abdominal CT scans
Raffaella Fiamma Cabini1,2, Andrea Cozzi3, Svenja Leu3
1Euler Institute, Università della Svizzera italiana, Lugano, Switzerland.
European Radiology Experimental
|January 29, 2025
Summary
CompositIA, an AI tool, automates body composition scoring from CT scans, reducing manual errors and time. This open-source pipeline achieves high precision, simplifying clinical assessments.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- Manual body composition scoring from CT scans is laborious and error-prone.
- Accurate body composition analysis is crucial for understanding tissue volume and physical properties.
Purpose of the Study:
- To develop and validate CompositIA, an automated, open-source pipeline for quantifying body composition scores from thoraco-abdominal CT scans.
- To improve the efficiency and accuracy of body composition analysis in clinical settings.
Main Methods:
- Trained CompositIA on 205 CT scans and validated on 54 independent scans.
- Utilized MultiResUNet for slice identification and U-nets for axial slice segmentation.
- Assessed performance using Mean Absolute Error (MAE), volumetric Dice Similarity Coefficient (vDSC), and relative error (PRE).
Main Results:
- CompositIA achieved MAE < 10 mm in 85% of cases for slice detection and vDSC > 0.85 for segmentation.
- Demonstrated strong agreement between automated and manual scores (p < 0.001) with mean PREs ranging from 5.13% to 15.18%.
Conclusions:
- CompositIA enables precise, automated quantification of body composition scores from CT scans.
- This open-source pipeline simplifies clinical assessments and broadens the application of body composition analysis.
