Flywheel gear that implements CaPTk's application for single-subject radiomic feature extraction. Gear will output a CSV file with the subject's radiomic features within the tumor mask, for all of the input scans.
NOTE: input images must be co-registered
Normalization (to range 0-255) is an optional step.
- CaPTk (gear uses the existing Docker container 2021.03.29)
- Tumor segmentation mask (gear automatically binarizes input mask - all non-zero voxels become 1)
- T1
- T1CE
- T2
- FLAIR
- ADC
- Radiomic feature parameter file (if not specified, gear will use CaPTk's default 3D parameters)
- brain mask (for normalization step)
- subject label to use in the output file
- output file name (defaults to radiomic_features.csv)
- perform normalization (for T1/T1CE/T2/FLAIR inputs)
- use intersection of tumor segmentation & brain mask during normalization
0.1.0 - Added optional normalization step with associated configurations