def run_main():
sct.init_sct()
parser = get_parser()
args = sys.argv[1:]
arguments = parser.parse(args)
# Input filename
fname_input_data = arguments["-i"]
fname_data = os.path.abspath(fname_input_data)
# Method used
method = 'optic'
if "-method" in arguments:
method = arguments["-method"]
# Contrast type
contrast_type = ''
if "-c" in arguments:
contrast_type = arguments["-c"]
if method == 'optic' and not contrast_type:
# Contrast must be
error = 'ERROR: -c is a mandatory argument when using Optic method.'
sct.printv(error, type='error')
return
# Ga between slices
interslice_gap = 10.0
if "-gap" in arguments:
interslice_gap = float(arguments["-gap"])
# Output folder
if "-o" in arguments:
file_output = arguments["-o"]
else:
path_data, file_data, ext_data = sct.extract_fname(fname_data)
file_output = os.path.join(path_data, file_data + '_centerline')
# Verbosity
verbose = 0
if "-v" in arguments:
verbose = int(arguments["-v"])
if method == 'viewer':
im_labels = _call_viewer_centerline(Image(fname_data),
interslice_gap=interslice_gap)
im_centerline, arr_centerline, _ = \
get_centerline(im_labels, algo_fitting='polyfit', param=ParamCenterline(degree=3), minmax=True,
verbose=verbose)
else:
im_centerline, arr_centerline, _ = \
get_centerline(Image(fname_data), algo_fitting='optic', param=ParamCenterline(contrast=contrast_type),
minmax=True, verbose=verbose)
# save centerline as nifti (discrete) and csv (continuous) files
im_centerline.save(file_output + '.nii.gz')
np.savetxt(file_output + '.csv', arr_centerline.transpose(), delimiter=",")
sct.display_viewer_syntax([fname_input_data, file_output + '.nii.gz'],
colormaps=['gray', 'red'],
opacities=['', '1'])
def run_main():
sct.init_sct()
parser = get_parser()
args = sys.argv[1:]
arguments = parser.parse(args)
# Input filename
fname_input_data = arguments["-i"]
fname_data = os.path.abspath(fname_input_data)
# Method used
method = 'optic'
if "-method" in arguments:
method = arguments["-method"]
# Contrast type
contrast_type = ''
if "-c" in arguments:
contrast_type = arguments["-c"]
if method == 'optic' and not contrast_type:
# Contrast must be
error = 'ERROR: -c is a mandatory argument when using Optic method.'
sct.printv(error, type='error')
return
# Ga between slices
interslice_gap = 10.0
if "-gap" in arguments:
interslice_gap = float(arguments["-gap"])
# Output folder
if "-o" in arguments:
file_output = arguments["-o"]
else:
path_data, file_data, ext_data = sct.extract_fname(fname_data)
file_output = os.path.join(path_data, file_data + '_centerline')
verbose = int(arguments.get('-v'))
sct.init_sct(log_level=verbose, update=True) # Update log level
if method == 'viewer':
im_labels = _call_viewer_centerline(Image(fname_data), interslice_gap=interslice_gap)
im_centerline, arr_centerline, _ = \
get_centerline(im_labels, algo_fitting='polyfit', degree=3, minmax=True, verbose=verbose)
else:
im_centerline, arr_centerline, _ = \
get_centerline(Image(fname_data), algo_fitting='optic', contrast=contrast_type, minmax=True,
verbose=verbose)
# save centerline as nifti (discrete) and csv (continuous) files
im_centerline.save(file_output + '.nii.gz')
np.savetxt(file_output + '.csv', arr_centerline.transpose(), delimiter=",")
sct.display_viewer_syntax([fname_input_data, file_output+'.nii.gz'], colormaps=['gray', 'red'], opacities=['', '1'])
def find_centerline(algo, image_fname, contrast_type, brain_bool,
folder_output, remove_temp_files, centerline_fname):
"""
Assumes RPI orientation
:param algo:
:param image_fname:
:param contrast_type:
:param brain_bool:
:param folder_output:
:param remove_temp_files:
:param centerline_fname:
:return:
"""
# TODO: remove unnecessary i/o
if Image(image_fname).dim[2] == 1: # isct_spine_detect requires nz > 1
im_concat = concat_data([image_fname, image_fname], dim=2)
im_concat.save(sct.add_suffix(image_fname, '_concat'))
image_fname = sct.add_suffix(image_fname, '_concat')
bool_2d = True
else:
bool_2d = False
# TODO: maybe change 'svm' for 'optic', because this is how we call it in sct_get_centerline
if algo == 'svm':
# run optic on a heatmap computed by a trained SVM+HoG algorithm
# optic_models_fname = os.path.join(path_sct, 'data', 'optic_models', '{}_model'.format(contrast_type))
# # TODO: replace with get_centerline(method=optic)
img_ctl, arr_ctl, _ = get_centerline(Image(image_fname),
algo_fitting='optic',
contrast=contrast_type)
centerline_filename = sct.add_suffix(image_fname, "_ctr")
img_ctl.save(centerline_filename)
elif algo == 'cnn':
# CNN parameters
dct_patch_ctr = {
't2': {
'size': (80, 80),
'mean': 51.1417,
'std': 57.4408
},
't2s': {
'size': (80, 80),
'mean': 68.8591,
'std': 71.4659
},
't1': {
'size': (80, 80),
'mean': 55.7359,
'std': 64.3149
},
'dwi': {
'size': (80, 80),
'mean': 55.744,
'std': 45.003
}
}
dct_params_ctr = {
't2': {
'features': 16,
'dilation_layers': 2
},
't2s': {
'features': 8,
'dilation_layers': 3
},
't1': {
'features': 24,
'dilation_layers': 3
},
'dwi': {
'features': 8,
'dilation_layers': 2
}
}
# load model
ctr_model_fname = os.path.join(sct.__sct_dir__, 'data',
'deepseg_sc_models',
'{}_ctr.h5'.format(contrast_type))
ctr_model = nn_architecture_ctr(
height=dct_patch_ctr[contrast_type]['size'][0],
width=dct_patch_ctr[contrast_type]['size'][1],
channels=1,
classes=1,
features=dct_params_ctr[contrast_type]['features'],
depth=2,
temperature=1.0,
padding='same',
batchnorm=True,
dropout=0.0,
dilation_layers=dct_params_ctr[contrast_type]['dilation_layers'])
ctr_model.load_weights(ctr_model_fname)
logger.info("Resample the image to 0.5x0.5 mm in-plane resolution...")
fname_res = sct.add_suffix(image_fname, '_resampled')
input_resolution = Image(image_fname).dim[4:7]
new_resolution = 'x'.join(['0.5', '0.5', str(input_resolution[2])])
resampling.resample_file(image_fname,
fname_res,
new_resolution,
'mm',
'linear',
verbose=0)
# compute the heatmap
fname_heatmap = sct.add_suffix(image_fname, "_heatmap")
img_filename = ''.join(sct.extract_fname(fname_heatmap)[:2])
fname_heatmap_nii = img_filename + '.nii'
z_max = heatmap(filename_in=fname_res,
filename_out=fname_heatmap_nii,
model=ctr_model,
patch_shape=dct_patch_ctr[contrast_type]['size'],
mean_train=dct_patch_ctr[contrast_type]['mean'],
std_train=dct_patch_ctr[contrast_type]['std'],
brain_bool=brain_bool)
# run optic on the heatmap
centerline_filename = sct.add_suffix(fname_heatmap, "_ctr")
heatmap2optic(fname_heatmap=fname_heatmap_nii,
lambda_value=7 if contrast_type == 't2s' else 1,
fname_out=centerline_filename,
z_max=z_max if brain_bool else None)
elif algo == 'viewer':
im_labels = _call_viewer_centerline(Image(image_fname))
im_centerline, arr_centerline, _ = get_centerline(im_labels)
centerline_filename = sct.add_suffix(image_fname, "_ctr")
labels_filename = sct.add_suffix(image_fname, "_labels-centerline")
im_centerline.save(centerline_filename)
im_labels.save(labels_filename)
elif algo == 'file':
centerline_filename = sct.add_suffix(image_fname, "_ctr")
# Re-orient the manual centerline
Image(centerline_fname).change_orientation('RPI').save(
centerline_filename)
else:
logger.error(
'The parameter "-centerline" is incorrect. Please try again.')
sys.exit(1)
if algo != 'cnn':
logger.info("Resample the image to 0.5x0.5 mm in-plane resolution...")
fname_res = sct.add_suffix(image_fname, '_resampled')
input_resolution = Image(image_fname).dim[4:7]
new_resolution = 'x'.join(['0.5', '0.5', str(input_resolution[2])])
resampling.resample_file(image_fname,
fname_res,
new_resolution,
'mm',
'linear',
verbose=0)
resampling.resample_file(centerline_filename,
centerline_filename,
new_resolution,
'mm',
'linear',
verbose=0)
if bool_2d:
im_split_lst = split_data(Image(centerline_filename), dim=2)
im_split_lst[0].save(centerline_filename)
return fname_res, centerline_filename
def find_centerline(algo, image_fname, contrast_type, brain_bool, folder_output, remove_temp_files, centerline_fname):
"""
Assumes RPI orientation
:param algo:
:param image_fname:
:param contrast_type:
:param brain_bool:
:param folder_output:
:param remove_temp_files:
:param centerline_fname:
:return:
"""
# TODO: remove unnecessary i/o
if Image(image_fname).dim[2] == 1: # isct_spine_detect requires nz > 1
im_concat = concat_data([image_fname, image_fname], dim=2)
im_concat.save(sct.add_suffix(image_fname, '_concat'))
image_fname = sct.add_suffix(image_fname, '_concat')
bool_2d = True
else:
bool_2d = False
# TODO: maybe change 'svm' for 'optic', because this is how we call it in sct_get_centerline
if algo == 'svm':
# run optic on a heatmap computed by a trained SVM+HoG algorithm
# optic_models_fname = os.path.join(path_sct, 'data', 'optic_models', '{}_model'.format(contrast_type))
# # TODO: replace with get_centerline(method=optic)
img_ctl, arr_ctl, _ = get_centerline(Image(image_fname), algo_fitting='optic', contrast=contrast_type)
centerline_filename = sct.add_suffix(image_fname, "_ctr")
img_ctl.save(centerline_filename)
elif algo == 'cnn':
# CNN parameters
dct_patch_ctr = {'t2': {'size': (80, 80), 'mean': 51.1417, 'std': 57.4408},
't2s': {'size': (80, 80), 'mean': 68.8591, 'std': 71.4659},
't1': {'size': (80, 80), 'mean': 55.7359, 'std': 64.3149},
'dwi': {'size': (80, 80), 'mean': 55.744, 'std': 45.003}}
dct_params_ctr = {'t2': {'features': 16, 'dilation_layers': 2},
't2s': {'features': 8, 'dilation_layers': 3},
't1': {'features': 24, 'dilation_layers': 3},
'dwi': {'features': 8, 'dilation_layers': 2}}
# load model
ctr_model_fname = os.path.join(sct.__sct_dir__, 'data', 'deepseg_sc_models', '{}_ctr.h5'.format(contrast_type))
ctr_model = nn_architecture_ctr(height=dct_patch_ctr[contrast_type]['size'][0],
width=dct_patch_ctr[contrast_type]['size'][1],
channels=1,
classes=1,
features=dct_params_ctr[contrast_type]['features'],
depth=2,
temperature=1.0,
padding='same',
batchnorm=True,
dropout=0.0,
dilation_layers=dct_params_ctr[contrast_type]['dilation_layers'])
ctr_model.load_weights(ctr_model_fname)
logger.info("Resample the image to 0.5x0.5 mm in-plane resolution...")
fname_res = sct.add_suffix(image_fname, '_resampled')
input_resolution = Image(image_fname).dim[4:7]
new_resolution = 'x'.join(['0.5', '0.5', str(input_resolution[2])])
resampling.resample_file(image_fname, fname_res, new_resolution, 'mm', 'linear', verbose=0)
# compute the heatmap
fname_heatmap = sct.add_suffix(image_fname, "_heatmap")
img_filename = ''.join(sct.extract_fname(fname_heatmap)[:2])
fname_heatmap_nii = img_filename + '.nii'
z_max = heatmap(filename_in=fname_res,
filename_out=fname_heatmap_nii,
model=ctr_model,
patch_shape=dct_patch_ctr[contrast_type]['size'],
mean_train=dct_patch_ctr[contrast_type]['mean'],
std_train=dct_patch_ctr[contrast_type]['std'],
brain_bool=brain_bool)
# run optic on the heatmap
centerline_filename = sct.add_suffix(fname_heatmap, "_ctr")
heatmap2optic(fname_heatmap=fname_heatmap_nii,
lambda_value=7 if contrast_type == 't2s' else 1,
fname_out=centerline_filename,
z_max=z_max if brain_bool else None)
elif algo == 'viewer':
im_labels = _call_viewer_centerline(Image(image_fname))
im_centerline, arr_centerline, _ = get_centerline(im_labels)
centerline_filename = sct.add_suffix(image_fname, "_ctr")
labels_filename = sct.add_suffix(image_fname, "_labels-centerline")
im_centerline.save(centerline_filename)
im_labels.save(labels_filename)
elif algo == 'file':
centerline_filename = sct.add_suffix(image_fname, "_ctr")
# Re-orient the manual centerline
Image(centerline_fname).change_orientation('RPI').save(centerline_filename)
else:
logger.error('The parameter "-centerline" is incorrect. Please try again.')
sys.exit(1)
if algo != 'cnn':
logger.info("Resample the image to 0.5x0.5 mm in-plane resolution...")
fname_res = sct.add_suffix(image_fname, '_resampled')
input_resolution = Image(image_fname).dim[4:7]
new_resolution = 'x'.join(['0.5', '0.5', str(input_resolution[2])])
resampling.resample_file(image_fname, fname_res, new_resolution, 'mm', 'linear', verbose=0)
resampling.resample_file(centerline_filename, centerline_filename, new_resolution, 'mm', 'linear', verbose=0)
if bool_2d:
im_split_lst = split_data(Image(centerline_filename), dim=2)
im_split_lst[0].save(centerline_filename)
return fname_res, centerline_filename
def run_main():
sct.init_sct()
parser = get_parser()
args = sys.argv[1:]
arguments = parser.parse(args)
# Input filename
fname_input_data = arguments["-i"]
fname_data = os.path.abspath(fname_input_data)
# Method used
method = 'optic'
if "-method" in arguments:
method = arguments["-method"]
# Contrast type
contrast_type = ''
if "-c" in arguments:
contrast_type = arguments["-c"]
if method == 'optic' and not contrast_type:
# Contrast must be
error = 'ERROR: -c is a mandatory argument when using Optic method.'
sct.printv(error, type='error')
return
# Gap between slices
interslice_gap = 10.0
if "-gap" in arguments:
interslice_gap = float(arguments["-gap"])
param_centerline = ParamCenterline(
algo_fitting=arguments['-centerline-algo'],
smooth=arguments['-centerline-smooth'],
minmax=True)
# Output folder
if "-o" in arguments:
file_output = arguments["-o"]
else:
path_data, file_data, ext_data = sct.extract_fname(fname_data)
file_output = os.path.join(path_data, file_data + '_centerline')
verbose = int(arguments.get('-v'))
sct.init_sct(log_level=verbose, update=True) # Update log level
if method == 'viewer':
# Manual labeling of cord centerline
im_labels = _call_viewer_centerline(Image(fname_data), interslice_gap=interslice_gap)
else:
# Automatic detection of cord centerline
im_labels = Image(fname_data)
param_centerline.algo_fitting = 'optic'
param_centerline.contrast = contrast_type
# Extrapolate and regularize (or detect if optic) cord centerline
im_centerline, arr_centerline, _, _ = get_centerline(im_labels,
param=param_centerline,
verbose=verbose)
# save centerline as nifti (discrete) and csv (continuous) files
im_centerline.save(file_output + '.nii.gz')
np.savetxt(file_output + '.csv', arr_centerline.transpose(), delimiter=",")
sct.display_viewer_syntax([fname_input_data, file_output+'.nii.gz'], colormaps=['gray', 'red'], opacities=['', '1'])
def run_main():
init_sct()
parser = get_parser()
arguments = parser.parse_args(args=None if sys.argv[1:] else ['--help'])
# Input filename
fname_input_data = arguments.i
fname_data = os.path.abspath(fname_input_data)
# Method used
method = arguments.method
# Contrast type
contrast_type = arguments.c
if method == 'optic' and not contrast_type:
# Contrast must be
error = "ERROR: -c is a mandatory argument when using 'optic' method."
printv(error, type='error')
return
# Gap between slices
interslice_gap = arguments.gap
param_centerline = ParamCenterline(algo_fitting=arguments.centerline_algo,
smooth=arguments.centerline_smooth,
minmax=True)
# Output folder
if arguments.o is not None:
file_output = arguments.o
else:
path_data, file_data, ext_data = extract_fname(fname_data)
file_output = os.path.join(path_data, file_data + '_centerline')
verbose = int(arguments.v)
init_sct(log_level=verbose, update=True) # Update log level
if method == 'viewer':
# Manual labeling of cord centerline
im_labels = _call_viewer_centerline(Image(fname_data),
interslice_gap=interslice_gap)
elif method == 'fitseg':
im_labels = Image(fname_data)
elif method == 'optic':
# Automatic detection of cord centerline
im_labels = Image(fname_data)
param_centerline.algo_fitting = 'optic'
param_centerline.contrast = contrast_type
else:
printv(
"ERROR: The selected method is not available: {}. Please look at the help."
.format(method),
type='error')
return
# Extrapolate and regularize (or detect if optic) cord centerline
im_centerline, arr_centerline, _, _ = get_centerline(
im_labels, param=param_centerline, verbose=verbose)
# save centerline as nifti (discrete) and csv (continuous) files
im_centerline.save(file_output + '.nii.gz')
np.savetxt(file_output + '.csv', arr_centerline.transpose(), delimiter=",")
display_viewer_syntax([fname_input_data, file_output + '.nii.gz'],
colormaps=['gray', 'red'],
opacities=['', '1'])
path_qc = arguments.qc
qc_dataset = arguments.qc_dataset
qc_subject = arguments.qc_subject
# Generate QC report
if path_qc is not None:
generate_qc(fname_input_data,
fname_seg=file_output + '.nii.gz',
args=sys.argv[1:],
path_qc=os.path.abspath(path_qc),
dataset=qc_dataset,
subject=qc_subject,
process='sct_get_centerline')
display_viewer_syntax([fname_input_data, file_output + '.nii.gz'],
colormaps=['gray', 'red'],
opacities=['', '0.7'])
def find_centerline(algo, image_fname, contrast_type, brain_bool, folder_output, remove_temp_files, centerline_fname):
"""
Assumes RPI orientation
:param algo:
:param image_fname:
:param contrast_type:
:param brain_bool:
:param folder_output:
:param remove_temp_files:
:param centerline_fname:
:return:
"""
im = Image(image_fname)
ctl_absolute_path = sct.add_suffix(im.absolutepath, "_ctr")
# isct_spine_detect requires nz > 1
if im.dim[2] == 1:
im = concat_data([im, im], dim=2)
im.hdr['dim'][3] = 2 # Needs to be change manually since dim not updated during concat_data
bool_2d = True
else:
bool_2d = False
# TODO: maybe change 'svm' for 'optic', because this is how we call it in sct_get_centerline
if algo == 'svm':
# run optic on a heatmap computed by a trained SVM+HoG algorithm
# optic_models_fname = os.path.join(path_sct, 'data', 'optic_models', '{}_model'.format(contrast_type))
# # TODO: replace with get_centerline(method=optic)
im_ctl, _, _, _ = get_centerline(im,
ParamCenterline(algo_fitting='optic', contrast=contrast_type))
elif algo == 'cnn':
# CNN parameters
dct_patch_ctr = {'t2': {'size': (80, 80), 'mean': 51.1417, 'std': 57.4408},
't2s': {'size': (80, 80), 'mean': 68.8591, 'std': 71.4659},
't1': {'size': (80, 80), 'mean': 55.7359, 'std': 64.3149},
'dwi': {'size': (80, 80), 'mean': 55.744, 'std': 45.003}}
dct_params_ctr = {'t2': {'features': 16, 'dilation_layers': 2},
't2s': {'features': 8, 'dilation_layers': 3},
't1': {'features': 24, 'dilation_layers': 3},
'dwi': {'features': 8, 'dilation_layers': 2}}
# load model
ctr_model_fname = os.path.join(sct.__sct_dir__, 'data', 'deepseg_sc_models', '{}_ctr.h5'.format(contrast_type))
ctr_model = nn_architecture_ctr(height=dct_patch_ctr[contrast_type]['size'][0],
width=dct_patch_ctr[contrast_type]['size'][1],
channels=1,
classes=1,
features=dct_params_ctr[contrast_type]['features'],
depth=2,
temperature=1.0,
padding='same',
batchnorm=True,
dropout=0.0,
dilation_layers=dct_params_ctr[contrast_type]['dilation_layers'])
ctr_model.load_weights(ctr_model_fname)
# compute the heatmap
im_heatmap, z_max = heatmap(im=im,
model=ctr_model,
patch_shape=dct_patch_ctr[contrast_type]['size'],
mean_train=dct_patch_ctr[contrast_type]['mean'],
std_train=dct_patch_ctr[contrast_type]['std'],
brain_bool=brain_bool)
im_ctl, _, _, _ = get_centerline(im_heatmap,
ParamCenterline(algo_fitting='optic', contrast=contrast_type))
if z_max is not None:
sct.printv('Cropping brain section.')
im_ctl.data[:, :, z_max:] = 0
elif algo == 'viewer':
im_labels = _call_viewer_centerline(im)
im_ctl, _, _, _ = get_centerline(im_labels, param=ParamCenterline())
elif algo == 'file':
im_ctl = Image(centerline_fname)
im_ctl.change_orientation('RPI')
else:
logger.error('The parameter "-centerline" is incorrect. Please try again.')
sys.exit(1)
# TODO: for some reason, when algo == 'file', the absolutepath is changed to None out of the method find_centerline
im_ctl.absolutepath = ctl_absolute_path
if bool_2d:
im_ctl = split_data(im_ctl, dim=2)[0]
if algo != 'viewer':
im_labels = None
# TODO: remove unecessary return params
return "dummy_file_name", im_ctl, im_labels
