Limited capability of MRI radiomics to predict primary tumor histology of brain metastases in external validation

URN to cite this document:: urn:nbn:de:bvb:355-epub-584215
DOI to cite this document:: 10.5283/epub.58421

Strotzer, Quirin David

; Wagner, Thomas

; Angstwurm, Pia ; Hense, Katharina

; Scheuermeyer, Lucca ; Noeva, Ekaterina ; Dinkel, Johannes ; Stroszczynski, Christian

; Fellner, Claudia

; Riemenschneider, Markus J.

; Rosengarth, Katharina

; Pukrop, Tobias

; Wiesinger, Isabel

; Wendl, Christina ; Schicho, Andreas

License: Creative Commons Attribution 4.0
PDF - Published Version
(12MB)

Date of publication of this fulltext: 12 Jun 2024 10:33

Alternative links to fulltext:DOI Publisher

Details

Preview

Export bibliographical data

Abstract

Abstract

Background

Growing research demonstrates the ability to predict histology or genetic information of various malignancies using radiomic features extracted from imaging data. This study aimed to investigate MRI-based radiomics in predicting the primary tumor of brain metastases through internal and external validation, using oversampling techniques to address the class imbalance.
Methods

This IRB-approved retrospective multicenter study included brain metastases from lung cancer, melanoma, breast cancer, colorectal cancer, and a combined heterogenous group of other primary entities (5-class classification). Local data were acquired between 2003 and 2021 from 231 patients (545 metastases). External validation was performed with 82 patients (280 metastases) and 258 patients (809 metastases) from the publicly available Stanford BrainMetShare and the University of California San Francisco Brain Metastases Stereotactic Radiosurgery datasets, respectively. Preprocessing included brain extraction, bias correction, coregistration, intensity normalization, and semi-manual binary tumor segmentation. Two-thousand five hundred and twenty-eight radiomic features were extracted from T1w (± contrast), fluid-attenuated inversion recovery (FLAIR), and wavelet transforms for each sequence (8 decompositions). Random forest classifiers were trained with selected features on original and oversampled data (5-fold cross-validation) and evaluated on internal/external holdout test sets using accuracy, precision, recall, F1 score, and area under the receiver-operating characteristic curve (AUC).
Results

Oversampling did not improve the overall unsatisfactory performance on the internal and external test sets. Incorrect data partitioning (oversampling before train/validation/test split) leads to a massive overestimation of model performance.
Conclusions

Radiomics models’ capability to predict histologic or genomic data from imaging should be critically assessed; external validation is essential.

Owner only: item control page

Download Statistics

Altmetric

Alternative Statistics (altmetrics)

More literature (via CORE)

Details

Preview

Export bibliographical data

Abstract

Abstract

Download Statistics

Downloads

Alternative Statistics (altmetrics)

More literature (via CORE)

University Library