<b>Multimodal MRI radiomics</b><b> based on </b><b>habitat subregions of the tumor microenvironment</b><b> for predicting risk stratification in glioblastoma</b>

This dataset contains the minimal anonymized data necessary to replicate the key findings presented in the manuscript titled "Multimodal MRI radiomics based on habitat subregions of the tumor microenvironment for predicting risk stratification in glioblastoma" (Han Wang, Plos One, 2025).The data comprises three core components essential for the study's analyses:1. Clinical Data (clinical.csv):Anonymized demographic and clinical variables for 423 patients/subjects.Includes variables such as: age, gender, GTR_over90percent.Note: All direct identifiers (e.g., patient ID, name, address, exact dates of birth) have been removed or generalized (e.g., age groups) to ensure participant anonymity.2. Radiomics Features (ED.csv, ET.csv, NC.csv, ED+ET+NC.csv):**Quantitative imaging features extracted from MRI scans of T1WI-CE, FLAIR, and DTI-FA.A fully automated approach involving label fusion from multiple deep learning algorithms was used to segment distinct tumor subregions histologically. The segmentation focused primarily on delineating the enhanced (ET), necrotic core (NC), and edema (ED) regions. All the segmentation results were manually reviewed and corrected using ITK-SNAP to ensure accuracy. Following segmentation, quantitative imaging phenomic (QIP) features were derived from each tumor subregion with the Cancer Imaging Phenomics Toolkit (CaPTk) in accordance with the guidelines established by the Image Biomarker Standardisation Initiative (IBSI).A total of 145 features were extracted for each subregion, encompassing five categories:(1) Intensity-based features, or first-order statistics (e.g., mean, median, maximum, minimum, standard deviation, skewness, kurtosis).(2) Histogram-related features, which describe the range and distribution of gray-level intensities.(3) Volumetric measurements, including shape such as elongation, perimeter, principal component axes, and area/volume for 2D/3D data.(4) Morphological parameters, reflecting the geometric properties of the tumor.(5) Textural descriptors, including indices derived from the gray-level co-occurrence matrix (GLCM), gray-level run-length matrix (GLRLM), gray-level size zone matrix (GLSZM), neighborhood gray-tone difference matrix (NGTDM), and local binary pattern (LBP).3. Label Data (label.csv and survival.csv):**The cohort was randomly divided into training and testing cohorts at a 7:3 ratio..Ethics and Anonymization:The dataset used in this study was obtained from The Cancer Imaging Archive (TCIA). The use of the public database complied with the citation requirements and data use policies listed on the public portal of the TCGA-TCIA website. The study was exempt from institutional review board review and approval because patient identifiers were not available to database users.