How Robust Are Prognostic Radiomic Features? Comparing Selection Techniques on Pooled Glioblastoma Cohorts
Abstract
Radiomic survival studies in glioblastoma report heterogeneous prognostic features, but how much of this stems from the methodology itself is rarely examined. We analysed the association of 463 radiomic, anatomical and clinical features from pre-treatment MRI with overall survival in 705 IDH-wildtype glioblastoma patients from three public cohorts, using three techniques: a log-rank test after median split, univariate Cox regression (both Benjamini-Hochberg corrected) and an XGBoost regressor interpreted with SHAP. The log-rank test selected 99 features, Cox regression 72 and SHAP its top 20; five were selected by all three: age, a peritumoural T2 texture feature, the number of connected peritumoural mask components and two image-level intensity maxima. The two survival tests shared 51 features, about half of the SHAP top 20 were also selected by either test, and where selections overlapped, effect directions agreed. Many selections were PyRadiomics diagnostic descriptors that are not restricted to tumour voxels. Analytical pipeline choices, including ranking method, pre-filtering and descriptor handling, can substantially influence which features appear prognostic.
Related
- Prognostic Value of Combined Baseline FDG PET/CT Radiomic Parameters in Primary Mediastinal B-Cell Lymphoma
- Comparing covariate selection strategies for predictive dynamic occupancy models
- GSC State Plasticity Drives Therapeutic Resistance and Recurrence in Glioblastoma
- Lactate Promotes Tunneling Nanotube Formation and Reduces Temozolomide Sensitivity in Glioblastoma Cells
- Cascaded Dual-Targeting (WTW/TAT) Liposomes Delivery DOX for Precision Glioblastoma Treatment
BibTeX
@misc{sodonougbo2026robust,
title = {{How Robust Are Prognostic Radiomic Features? Comparing Selection Techniques on Pooled Glioblastoma Cohorts}},
author = {Yannick Sodonougbo and Marlene Schnieders and Alexia Rizoudis and Santiago Cepeda and Frank Kramer and Dominik Müller},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.26095},
eprint = {2610.26095},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.26095},
}
RIS
TY - UNPB
TI - How Robust Are Prognostic Radiomic Features? Comparing Selection Techniques on Pooled Glioblastoma Cohorts
AU - Yannick Sodonougbo
AU - Marlene Schnieders
AU - Alexia Rizoudis
AU - Santiago Cepeda
AU - Frank Kramer
AU - Dominik Müller
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.26095
UR - https://arxiv.org/abs/2610.26095
AB - Radiomic survival studies in glioblastoma report heterogeneous prognostic features, but how much of this stems from the methodology itself is rarely examined. We analysed the association of 463 radiomic, anatomical and clinical features from pre-treatment MRI with overall survival in 705 IDH-wildtype glioblastoma patients from three public cohorts, using three techniques: a log-rank test after median split, univariate Cox regression (both Benjamini-Hochberg corrected) and an XGBoost regressor interpreted with SHAP. The log-rank test selected 99 features, Cox regression 72 and SHAP its top 20; five were selected by all three: age, a peritumoural T2 texture feature, the number of connected peritumoural mask components and two image-level intensity maxima. The two survival tests shared 51 features, about half of the SHAP top 20 were also selected by either test, and where selections overlapped, effect directions agreed. Many selections were PyRadiomics diagnostic descriptors that are not restricted to tumour voxels. Analytical pipeline choices, including ranking method, pre-filtering and descriptor handling, can substantially influence which features appear prognostic.
ER -