Enhancing Radiomic Feature Selection in PET-Based Heterogeneity Analysis through Tumor-Volume-Weighted Normalization
Description
This dataset contains resources developed to support the study of tumor heterogeneity and tumor volume-weighted normalization (TVWN) for radiomic feature analysis. The dataset which was developed using Python (version 3.11.0), enables reproducibility of radiomic feature extraction and calculations and provides tools for exploring and visualizing tumor heterogeneity which has been described in our work. Components: Simulated Tumor Images: The dataset includes computer-simulated tumor images provided in DICOM format. These images represent varying levels of tumor heterogeneity and noise levels. Python Scripts: Heterogeneous Tumor Simulation.py: A Python script designed for the simulation and visualization of tumor heterogeneity. Tumor-Volume-Weighted Normalization Analysis.py: A Python script that implements tumor-volume-weighted normalization for radiomic feature standardization. Both scripts are essential for replicating the methodologies described in this work and are available for download as part of the dataset. Purpose: It provides tools to: Simulate tumor heterogeneity. Perform tumor-volume-weighted normalization of radiomic features. Advance reproducibility and validation efforts in radiomic feature extraction and analysis.