/bodynavigation

Body navigation in CT images

Primary LanguageJupyter NotebookMIT LicenseMIT

bodynavigation

Segmentation of internal organs from Computed Tomography images of- ten uses intensity and shape properties. We introduce a navigation system based on robust segmentation of body tissues like spine, body surface and lungs. Pose esti- mation of an investigated tissue can be performed using this algorithm and it also can be used as a support information for various segmentation algorithms. Preci- sion of liver segmentation based on Bayes classifier is shown in this paper and it is compared with state of the art methods using SLIVER07 dataset.

Spine rotation spine_rotation Diaphragm segmentation diaphragm

Install

conda install -c mjirik bodynavigation

Example

import io3d
import sed3

import bodynavigation

data3d, metadata = io3d.read("dicomdir/")

ss = bodynavigation.body_navigation.BodyNavigation(data3d, metadata["voxelsize_mm"])
seg = ss.get_diaphragm_mask().astype(np.uint8)
sed3.show_slices(data3d, seg*2, slice_step=20, axis=1, flipV=True)

Simple example can be found in examples directory.

Usefull API functions

ss = bodynavigation.body_navigation.BodyNavigation(data3d, metadata["voxelsize_mm"])

dsag = ss.dist_sagittal()
dcor = ss.dist_coronal()
daxi = ss.dist_axial()
ddia = ss.dist_diaphragm()
dsur = ss.dist_to_surface()
dspi = ss.dist_to_spine()

Conda build

conda build . -c conda-forge -c simpleitk -c bioconda --python 3.6