ANTs visualization module

Author: armaneshaghi

nipype/interfaces/ants/__init__.py

@@ -13,5 +13,8 @@
 # Segmentation Programs
 from .segmentation import Atropos, LaplacianThickness, N4BiasFieldCorrection
 
+# Visualization Programs
+from .visualization import ConvertScalarImageToRGB
+
 # Utility Programs
 from .utils import AverageAffineTransform, AverageImages, MultiplyImages, JacobianDeterminant

nipype/interfaces/ants/resampling.py

@@ -1,5 +1,3 @@
-# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
-# vi: set ft = python sts = 4 ts = 4 sw = 4 et:
 """ANTS Apply Transforms interface
 
    Change directory to provide relative paths for doctests

nipype/interfaces/ants/utils.py

@@ -1,5 +1,3 @@
-# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
-# vi: set ft = python sts = 4 ts = 4 sw = 4 et:
 """ANTS Apply Transforms interface
 
    Change directory to provide relative paths for doctests

nipype/interfaces/ants/visualization.py

@@ -0,0 +1,150 @@
+"""The ants visualisation module provides basic functions based on ITK.
+   Change directory to provide relative paths for doctests
+   >>> import os
+   >>> filepath = os.path.dirname( os.path.realpath( __file__ ) )
+   >>> datadir = os.path.realpath(os.path.join(filepath, '../../testing/data'))
+   >>> os.chdir(datadir)
+"""
+
+from ..base import (TraitedSpec, File, traits)
+from .base import ANTSCommand, ANTSCommandInputSpec
+import os
+from nipype.interfaces.base import InputMultiPath
+from nipype.interfaces.traits_extension import isdefined
+import numpy as np
+
+class ConvertScalarImageToRGBInputSpec(ANTSCommandInputSpec):
+    dimension=traits.Enum(3, 2, argstr= '%d', usedefault=True,
+                           desc='image dimension (2 or 3)', mandatory=True,
+                           position = 0)
+    input_image=File(argstr='%s', exists=True,
+            desc='Main input is a 3-D grayscale image.', mandatory=True,
+            position = 1)
+    output_image=traits.Str('rgb.nii.gz', argstr='%s', usedefault=True,
+                             desc=('rgb output image'), position = 2)
+    mask_image=File('none', argstr='%s', exists=True,
+                    desc = 'mask image', position = 3)
+    colormap=traits.Str(argstr='%s', usedefault=True,
+            desc=('Possible colormaps: grey, red, green, '
+                'blue, copper, jet, hsv, spring, summer, '
+                'autumn, winter, hot, cool, overunder, custom '
+                ), mandatory = True, position = 4)
+    custom_color_map_file=traits.Str('none', argstr='%s', usedefault=True, 
+            desc = 'custom color map file', position = 5)
+    minimum_input = traits.Int(argstr='%d', desc='minimum input',
+            mandatory = True, position = 6)
+    maximum_input = traits.Int(argstr='%d', desc='maximum input',
+            mandatory = True, position = 7)
+    minimum_RGB_output = traits.Int(0, usedefault=True,
+            argstr='%d', desc = '', position = 8)
+    maximum_RGB_output = traits.Int(255, usedefault=True,
+            argstr='%d', desc = '', position = 9)
+
+class ConvertScalarImageToRGBOutputSpec(TraitedSpec):
+    output_image= File(exists=True, desc='converted RGB image')
+
+class ConvertScalarImageToRGB(ANTSCommand):
+    """
+    Examples
+    --------
+    >>> from nipype.interfaces.ants.visualization import ConvertScalarImageToRGB
+    >>> converter = ConvertScalarImageToRGB()
+    >>> converter.inputs.dimension = 3
+    >>> converter.inputs.input_image = 'T1.nii.gz'
+    >>> converter.inputs.colormap = 'jet'
+    >>> converter.inputs.minimum_input = 0
+    >>> converter.inputs.maximum_input = 6 
+    >>> converter.cmdline
+    'ConvertScalarImageToRGB 3 T1.nii.gz rgb.nii.gz none jet none 0 6 0 255'
+    """
+    _cmd = 'ConvertScalarImageToRGB'
+    input_spec = ConvertScalarImageToRGBInputSpec
+    output_spec = ConvertScalarImageToRGBOutputSpec
+
+    def _format_arg(self, opt, spec, val):
+        return super(ConvertScalarImageToRGB, self)._format_arg(opt, spec, val)
+
+    def _list_outputs(self):
+        outputs = self._outputs().get()
+        outputs['output_image'] = os.path.join(os.getcwd(), 
+                self.inputs.output_image)
+        return outputs 
+
+
+class CreateTiledMosaicInputSpec(ANTSCommandInputSpec):
+    input_image = File(argstr='-i %s', exists=True,
+            desc = 'Main input is a 3-D grayscale image.',
+            mandatory = True)
+    rgb_image= File(argstr='-r %s', exists = True,
+            desc = ('An optional Rgb image can be added as an overlay.'
+            'It must have the same image' 
+            'geometry as the input grayscale image.'),
+            mandatory = True)
+    mask_image = File(argstr = '-x %s', exists = True,
+            desc = 'Specifies the ROI of the RGB voxels used.')
+    alpha_value = traits.Float(argstr = '-a %f',
+            desc = ('If an Rgb image is provided, render the overlay '
+            'using the specified alpha parameter.'))
+    output_image = traits.Str(argstr = '-o %s',
+            desc = 'The output consists of the tiled mosaic image.') 
+    tile_geometry = traits.Str(argstr = '%s',
+            desc = (
+          'The tile geometry specifies the number of rows and columns'
+          'in the output image. For example, if the user specifies ''5x10'', ' 
+          'then 5 rows by 10 columns of slices are rendered. If R < 0 and C > '
+          '0 (or vice versa), the negative value is selected'
+          'based on direction.') 
+    direction = traits.Int(argstr = '%d', desc = ('Specifies the direction of '
+          'the slices. If no direction is specified, the '
+          'direction with the coarsest spacing is chosen.')) 
+    pad_or_crop = traits.Str(argstr='-p %s', 
+        desc = 'argument passed to -p flag:'
+          '[padVoxelWidth,<constantValue=0>]'
+          '[lowerPadding[0]xlowerPadding[1],upperPadding[0]xupperPadding[1],'
+          'constantValue]'
+          'The user can specify whether to pad or crop a specified '
+          'voxel-width boundary of each individual slice. For this '
+          'program, cropping is simply padding with negative voxel-widths.' 
+          'If one pads (+), the user can also specify a constant pad '
+          'value (default = 0). If a mask is specified, the user can use '
+          'the mask to define the region, by using the keyword "mask"'
+          ' plus an offset, e.g. "-p mask+3".'
+        )
+    slices = traits.Str(argstr='-s %s',
+        desc = ('Number of slices to increment Slice1xSlice2xSlice3'
+        '[numberOfSlicesToIncrement,<minSlice=0>,<maxSlice=lastSlice>]'))
+    flip_slice = traits.Str(argstr = '-f %s',
+        desc = ('flipXxflipY'))
+    permute_axes = traits.Bool(argstr = '-g', desc = 'doPermute'
+        )
+    
+
+class CreateTiledMosaicOutputSpec(TraitedSpec):
+    output_image= File(exists=True, desc='image file')
+
+class CreateTiledMosaicOutputSpec(ANTSCommand):
+    """The program CreateTiledMosaic in conjunction with ConvertScalarImageToRGB
+    provides useful functionality for common image analysis tasks. The basic 
+    usage of CreateTiledMosaic is to tile a 3-D image volume slice-wise into 
+    a 2-D image.
+
+    Examples
+    --------
+
+    >>> from nipype.interfaces.ants.visualization import CreateTiledMosaic 
+    >>> mosaic_slicer = CreateTiledMosaic()
+    >>> mosaic_slicer.inputs.dimension = 3
+    >>> mosaic_slicer.inputs.input_image = 'T1.nii.gz'
+    >>> mosaic_slicer.inputs.colormap = 'jet'
+    >>>  
+
+    """
+
+    _cmd = 'CreateTiledMosaic'
+    input_spec = CreateTiledMosaicInputSpec
+    output_spec = CreateTiledMosaicInputSpecOutputSpec
+
+    def _list_outputs(self):
+        outputs = self._outputs().get()
+        outputs['output_image'] = os.path.join(os.getcwd(), 
+                self.inputs.output_image)