class FeatureSpatialOutputSpec(TraitedSpec):
edge_fraction = traits.Array(
desc=
"Array of the edge fraction feature scores for the components of the melIC file"
)
csf_fraction = traits.Array(
desc=
"Array of the CSF fraction feature scores for the components of the melIC file"
)
class TFRmorletInputSpec(BaseInterfaceInputSpec):
"""Input specification."""
epo_file = traits.File(exists=True,
desc='epochs file in .fif format',
mandatory=True)
freqs = traits.Array(exists=True, desc='frequencies in Hz', mandatory=True)
n_cycles = traits.Array(
desc='the number of cycles globally or for each frequency') # noqa
class ClassificationInputSpec(BaseInterfaceInputSpec):
maxRPcorr = traits.Array(
desc=
"Array of the 'maximum RP correlation' feature scores of the components"
)
edge_fraction = traits.Array(
desc="Array of the 'edge fraction' feature scores of the components")
HFC = traits.Array(
desc=
"Array of the 'high-frequency content' feature scores of the components"
)
csf_fraction = traits.Array(
desc="Array of the 'CSF fraction' feature scores of the components")
class CoherenceAnalyzerOutputSpec(TraitedSpec):
coherence_array = traits.Array(desc=('The pairwise coherence values'
'between the ROIs'))
timedelay_array = traits.Array(desc=('The pairwise time delays between the'
'ROIs (in seconds)'))
coherence_csv = File(desc=('A csv file containing the pairwise '
'coherence values'))
timedelay_csv = File(desc=('A csv file containing the pairwise '
'time delay values'))
coherence_fig = File(desc=('Figure representing coherence values'))
timedelay_fig = File(desc=('Figure representing coherence values'))
class LungSegmentationInferenceInputSpec(BaseInterfaceInputSpec):
tensor = traits.Array(desc='Tensor to be fed to the network.')
image_info = traits.Dict(
desc='Dictionary with information about the image.')
weights = traits.List(desc='List of network weights.')
outdir = Directory('segmented',
usedefault=True,
desc='Folder to store the preprocessing results.')
class ClassificationOutputSpec(TraitedSpec):
motionICs = traits.Array(
desc=
"Array containing the indices of the components identified as motion components"
)
classified_motion_ics = traits.File(
desc=
"A text file containing the indices of the components identified as motion components"
)
class PreprocessingOutputSpec(TraitedSpec):
preprocessed_array = traits.Array(
desc='preprocessed images (FLAIR only or T1 and FLAIR) as Numpy array')
slice_shape = traits.Tuple((traits.Int(), traits.Int(), traits.Int()),
desc='slice shape')
# ToDo: implement save npz
flair_array_npy = File(desc='preprocessed FLAIR as Numpy .npz')
class TrainSVMInputSpec(BaseInterfaceInputSpec):
inputs = File(exists=True,
desc='input values for training',
mandatory=True)
targets = File(exists=True,
desc='target values for training',
mandatory=True)
indices = traits.Array(desc='indices for inputs and targets',
mandatory=False)
class SignalPredictionInputSpec(BaseInterfaceInputSpec):
aligned_dwis = InputMultiObject(File(exists=True))
aligned_bvecs = traits.Either(InputMultiObject(File(exists=True)), traits.Array)
bvals = traits.Either(InputMultiObject(File(exists=True)), traits.Array)
aligned_mask = File(exists=True, mandatory=True)
aligned_b0_mean = File(exists=True, mandatory=True)
bvec_to_predict = traits.Array()
bval_to_predict = traits.Float()
minimal_q_distance = traits.Float(2.0, usedefault=True)
model = traits.Str('3dSHORE', usedefault=True)
class PredictInputSpec(BaseInterfaceInputSpec):
slice_shape = traits.Tuple((traits.Int(), traits.Int(), traits.Int()),
desc='slice shape')
preprocessed_array = traits.Array(
mandatory=True, desc='Array from preprocessed FLAIR and T1w')
weights = traits.List(File(exists=True, ),
mandatory=True,
desc='Weights as list of H5 files')
class _PlotFuncInputSpec(BaseInterfaceInputSpec):
"""Input interface wrapper for PlotFunc"""
conn_matrix = traits.Array(mandatory=True)
conn_model = traits.Str(mandatory=True)
atlas = traits.Any()
dir_path = Directory(exists=True, mandatory=True)
ID = traits.Any(mandatory=True)
network = traits.Any()
labels = traits.Array(mandatory=True)
roi = traits.Any()
coords = traits.Array(mandatory=True)
thr = traits.Any()
node_size = traits.Any()
edge_threshold = traits.Any()
smooth = traits.Any()
prune = traits.Any()
uatlas = traits.Any()
c_boot = traits.Any()
norm = traits.Any()
binary = traits.Bool()
hpass = traits.Any()
class PostprocessingInputSpec(BaseInterfaceInputSpec):
flair = File(exists=True,
mandatory=True,
desc='Input FLAIR in an ITK readable format')
prediction = traits.Array(mandatory=True, desc='prediction to reshape')
rows_standard = traits.Int(mandatory=True, desc='input size for rows')
cols_standard = traits.Int(mandatory=True, desc='input size for columns')
per = traits.Float(mandatory=True, desc='ratio')
class _PlotStructInputSpec(BaseInterfaceInputSpec):
"""Input interface wrapper for PlotStruct"""
conn_matrix = traits.Array(mandatory=True)
conn_model = traits.Str(mandatory=True)
atlas = traits.Any()
dir_path = Directory(exists=True, mandatory=True)
ID = traits.Any(mandatory=True)
network = traits.Any()
labels = traits.Array(mandatory=True)
roi = traits.Any()
coords = traits.Array(mandatory=True)
thr = traits.Any()
node_size = traits.Any()
edge_threshold = traits.Any()
prune = traits.Any()
uatlas = traits.Any()
target_samples = traits.Any()
norm = traits.Any()
binary = traits.Bool()
track_type = traits.Any()
directget = traits.Any()
min_length = traits.Any()
class _PlotStructInputSpec(BaseInterfaceInputSpec):
"""Input interface wrapper for PlotStruct"""
conn_matrix = traits.Any()
conn_model = traits.Str(mandatory=True)
atlas = traits.Any(mandatory=False)
dir_path = Directory(exists=True, mandatory=True)
ID = traits.Any(mandatory=True)
subnet = traits.Any(mandatory=True)
labels = traits.Array(mandatory=True)
roi = traits.Any(mandatory=True)
coords = traits.Array(mandatory=True)
thr = traits.Any(mandatory=True)
node_radius = traits.Any(mandatory=True)
edge_threshold = traits.Any(mandatory=True)
prune = traits.Any(mandatory=True)
parcellation = traits.Any(mandatory=False)
norm = traits.Any(mandatory=True)
binary = traits.Bool(mandatory=True)
track_type = traits.Any(mandatory=True)
traversal = traits.Any(mandatory=True)
min_length = traits.Any(mandatory=True)
error_margin = traits.Any(mandatory=True)
class _PlotFuncInputSpec(BaseInterfaceInputSpec):
"""Input interface wrapper for PlotFunc"""
conn_matrix = traits.Any()
conn_model = traits.Str(mandatory=True)
atlas = traits.Any(mandatory=False)
dir_path = Directory(exists=True, mandatory=True)
ID = traits.Any(mandatory=True)
subnet = traits.Any(mandatory=True)
labels = traits.Array(mandatory=True)
roi = traits.Any(mandatory=True)
coords = traits.Array(mandatory=True)
thr = traits.Any(mandatory=True)
node_radius = traits.Any(mandatory=True)
edge_threshold = traits.Any(mandatory=True)
smooth = traits.Any(mandatory=True)
prune = traits.Any(mandatory=True)
parcellation = traits.Any(mandatory=False)
norm = traits.Any(mandatory=True)
binary = traits.Bool(mandatory=True)
hpass = traits.Any(mandatory=True)
signal = traits.Any(mandatory=True)
edge_color_override = traits.Bool(mandatory=True)
class TrainInputSpec(BaseInterfaceInputSpec):
images = traits.Array(mandatory=True,
desc='Images for the training as NumPy array')
masks = traits.Array(mandatory=True,
desc='Masks for training as NumPy array')
model_path = traits.Directory(mandatory=True,
desc='directory where to save the models')
ensemble_parameter = traits.Int(3,
usedefault=True,
desc='ensemble parameter')
verbose = traits.Bool(True, usedefault=True, desc='Verbose')
batch_size = traits.Int(30, usedefault=True, desc='batch size, default 30')
epochs = traits.Int(5, usedefault=True, desc='epochs, default 5')
image_shape = traits.Tuple((traits.Int(), traits.Int(), traits.Int()),
desc='slice shape')
shuffle = traits.Bool(True, desc='shuffle')
class LungSegmentationPreprocOutputSpec(TraitedSpec):
preproc_image = traits.File(exists=True, desc='Preprocessed image')
tensor = traits.Array(desc='Tensor to be fed to the network.')
image_info = traits.Dict(
desc='Dictionary with information about the image.')
class PostprocessingOutputSpec(TraitedSpec):
postprocessed_prediction = traits.Array(
desc='get prediction in original shape')
class TestSVMInputSpec(BaseInterfaceInputSpec):
svm = File(exists=True, desc='SVM to be tested', mandatory=True)
inputs = traits.Array(desc='input values to be tested', mandatory=True)
targets = traits.Array(desc='target values to be tested', mandatory=True)
indices = traits.Array(desc='indices for inputs and targets',
mandatory=False)
class PredictOutputSpec(TraitedSpec):
prediction = traits.Array(desc='Prediction as array')
prediction_npy = traits.File(desc='Prediction as Numpy .npy')
class FeatureFrequencyOutputSpec(TraitedSpec):
HFC = traits.Array(
desc=
"Array of the HFC ('High-frequency content') feature scores for the components of the melodic_FTmix file"
)
class SavePredictionInputSpec(BaseInterfaceInputSpec):
prediction_array = traits.Array(mandatory=True, desc='prediction as array')
output_filename = traits.Str('prediction',
usedefault=True,
desc='output filename')
class EnsembleInputSpec(BaseInterfaceInputSpec):
in_arrays = traits.List(traits.Array(),
mandatory=True,
desc='Arrays list to esemble')
class FeatureTimeSeriesOutputSpec(TraitedSpec):
maxRPcorr = traits.Array(
desc=
"Array of the maximum RP correlation feature scores for the components of the melodic_mix file"
)
class EnsembleOutputSpec(TraitedSpec):
out_array = traits.Array(desc='averaged array')
class SplitB0DWIsFromFileOutputSpec(TraitedSpec):
out_B0s = OutputMultiPath(File(exists=True))
out_DWIs = OutputMultiPath(File(exists=True))
out_all = OutputMultiPath(File(exists=True))
out_indices = traits.Array(desc="B0 Indices in the table")
class ThresholdingInputSpec(BaseInterfaceInputSpec):
in_array = traits.Array(mandatory=True, desc='input array')
thres = traits.Float(default_value=0.5, usedefault=True, desc='threshold')
class WriteArrayToCsvInputSpec(BaseInterfaceInputSpec):
in_array = traits.Array(exists=True, mandatory=True, desc="array")
in_name = traits.String(mandatory=True, desc="Name of the output file")
class ExtractAffineOutputSpec(TraitedSpec):
out_matrix = traits.Array(desc='The affine matrix as NumPy array')
out_file = File(desc='A text file containing the affine matrix')
class ThresholdingOutputSpec(TraitedSpec):
out_array = traits.Array(desc='thresholded array')
