import os
import numpy as np
import nibabel as nb
import nighresjava
from ..io import load_volume, save_volume
from ..utils import _output_dir_4saving, _fname_4saving
[docs]def levelset_to_probability(levelset_image, distance_mm=5,
save_data=False, overwrite=False, output_dir=None,
file_name=None):
"""Levelset to probability
Creates a probability map from the distance to a levelset surface
representation.
Parameters
----------
levelset_image: niimg
Levelset image to be turned into probabilities
distance_mm: float, optional
Distance used for the range of probabilities in ]0,1[
save_data: bool, optional
Save output data to file (default is False)
overwrite: bool, optional
Overwrite existing results (default is False)
output_dir: str, optional
Path to desired output directory, will be created if it doesn't exist
file_name: str, optional
Desired base name for output files with file extension
(suffixes will be added)
Returns
----------
dict
Dictionary collecting outputs under the following keys
(suffix of output files in brackets)
* result (niimg): Probability map (output file suffix _l2p-proba)
Notes
----------
Ported from original Java module by Pierre-Louis Bazin
"""
print("\nLevelset to Probabilities")
# make sure that saving related parameters are correct
if save_data:
output_dir = _output_dir_4saving(output_dir, levelset_image)
proba_file = os.path.join(output_dir,
_fname_4saving(module=__name__,file_name=file_name,
rootfile=levelset_image,
suffix='l2p-proba'))
if overwrite is False \
and os.path.isfile(proba_file) :
print("skip computation (use existing results)")
output = {'result': proba_file}
return output
# load the data
lvl_img = load_volume(levelset_image)
lvl_data = lvl_img.get_data()
hdr = lvl_img.header
aff = lvl_img.affine
resolution = [x.item() for x in hdr.get_zooms()]
dimensions = lvl_data.shape
# algorithm
distance = distance_mm/np.min(resolution)
print("voxel spread: "+str(distance))
p_data = 0.5*(np.maximum(-1,np.minimum(1,-lvl_data/distance))+1)
hdr['cal_min'] = np.nanmin(p_data)
hdr['cal_max'] = np.nanmax(p_data)
proba = nb.Nifti1Image(p_data, aff, hdr)
if save_data:
save_volume(proba_file, proba)
return {'result': proba_file}
else:
return {'result': proba}