Expand description
Radiomics feature extractors (pure Rust).
First-order intensity statistics, Gray-Level Co-occurrence Matrix (GLCM) texture features at distance 1, and lite 2D/3D shape descriptors.
Modules§
- first_
order_ stats - First-order intensity statistics on a gray-level ROI.
- glcm_
features - Gray-Level Co-occurrence Matrix (GLCM) texture features at distance 1.
- shape_
2d - 2D shape features from a binary mask (compatibility wrapper).
- shape_
3d_ lite - Lite shape descriptors for binary masks (2D and sparse 3D voxel lists).
Structs§
- First
Order Stats - First-order radiomics features (IBSI-style subset).
- Glcm
Features - Haralick-style features averaged over 0° and 90° (symmetric GLCM).
- Shape2d
- Compact 2D shape summary.
- Shape2d
Features - 2D shape features from a binary mask (nonzero = foreground).
- Shape3d
Features - 3D shape features from a sparse voxel list.
Enums§
- Radiomics
Error - Shared radiomics error for the radiomics submodule.
Constants§
- DEFAULT_
HIST_ BINS - Default histogram bins for entropy when not specified.
- GLCM_
LEVELS_ 16 - Common quantization levels for radiomics GLCM.
- GLCM_
LEVELS_ 32 - GLCM_
MAX_ LEVELS - Max supported quantization levels (stack table size).
Functions§
- first_
order_ stats - Compute first-order stats over
samples(gray / intensity ROI values). - first_
order_ stats_ with_ bins - Same as
first_order_statswith explicit entropy histogram bin count. - glcm_
features - Compute GLCM features on a 2D gray image at distance 1 with 16 levels (default).
- glcm_
features_ d1 - Compute GLCM features on a 2D gray image at distance 1.
- shape_
2d_ features - Extract 2D shape from a binary mask (
nonzero= object). - shape_
2d_ from_ mask - Extract 2D shape from a binary mask (
nonzero= object). - shape_
3d_ from_ voxels - Extract 3D shape from a sparse list of voxel coordinates.