Revitalize the Potential of Radiomics: Interpretation and Feature Stability in Medical Imaging Analyses Through Groupwise Feature Importance
MCML Authors
Theresa Stüber
* Former Member
Michael Ingrisch
Prof. Dr.
Core PI
Abstract
Theresa Stüber
* Former Member
Michael Ingrisch
Prof. Dr.
Core PI
Abstract
Radiomics, involving analysis of calculated, quantitative features from medical images with machine learning tools, shares the instability challenge with other high-dimensional data analyses due to variations in the training set. This instability affects model interpretation and feature importance assessment. To enhance stability and interpretability, we introduce grouped feature importance, shedding light on tool limitations and advocating for more reliable radiomics-based analysis methods.
inproceedings SCI23
LB-D-DC 2023 @xAI 2023
Late-breaking Work, Demos and Doctoral Consortium at the 1st World Conference on eXplainable Artificial Intelligence. Lisbon, Portugal, Jul 26-28, 2023.Authors
A. T. Stüber • S. Coors • M. IngrischLinks
PDFResearch Area
BibTeXKey: SCI23