ABSTRACT
To date, there are almost no investigations addressing functional connectivity (FC) in patients with brain metastases (BM). In this retrospective study, we investigate the influence of BM on hemodynamic brain signals derived from functional magnetic resonance imaging (fMRI) and FC. Motor-fMRI data of 29 patients with BM and 29 matched healthy controls were analyzed to assess percent signal changes (PSC) in the ROIs motor cortex, premotor cortex, and supplementary motor cortex and FC in the sensorimotor, default mode, and salience networks using Statistical Parametric Mapping (SPM12) and marsbar and CONN toolboxes. In the PSC analysis, an attenuation of the BOLD signal in the metastases-affected hemisphere compared to the contralateral hemisphere was significant only in the supplementary motor cortex during hand movement. In the FC analysis, we found alterations in patients' FC compared to controls in all examined networks, also in the hemisphere contralateral to the metastasis. This indicates a qualitative attenuation of the BOLD signal in the affected hemisphere and also that FC is altered by the presence of BM, similarly to what is known for primary brain tumors. This transformation is not only visible in the infiltrated hemisphere, but also in the contralateral one, suggesting an influence of BM beyond local damage.
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 (±â 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.
