[ENH] Add new dipy interfaces

CHANGES

@@ -1,6 +1,8 @@
 Next release
 ============
 
+* ENH: New interfaces in dipy: RESTORE, EstimateResponseSH, CSD and StreamlineTractography
+       (https://github.com/nipy/nipype/pull/1090)
 * ENH: Added interfaces of AFNI (https://github.com/nipy/nipype/pull/1360,
        https://github.com/nipy/nipype/pull/1361)
 * ENH: Added support for PETPVC (https://github.com/nipy/nipype/pull/1335)

circle.yml

@@ -21,7 +21,7 @@ dependencies:
     # Set up python environment
     - pip install --upgrade pip
     - pip install -e .
-    - pip install matplotlib sphinx ipython boto coverage
+    - pip install matplotlib sphinx ipython boto coverage dipy
     - gem install fakes3
     - if [[ ! -d ~/examples/data ]]; then wget "http://tcpdiag.dl.sourceforge.net/project/nipy/nipype/nipype-0.2/nipype-tutorial.tar.bz2" && tar jxvf nipype-tutorial.tar.bz2 && mv nipype-tutorial/* ~/examples/; fi
     - if [[ ! -d ~/examples/fsl_course_data ]]; then wget -c "http://fsl.fmrib.ox.ac.uk/fslcourse/fdt1.tar.gz" && wget -c "http://fsl.fmrib.ox.ac.uk/fslcourse/fdt2.tar.gz" && wget -c "http://fsl.fmrib.ox.ac.uk/fslcourse/tbss.tar.gz" && mkdir ~/examples/fsl_course_data && tar zxvf fdt1.tar.gz -C ~/examples/fsl_course_data && tar zxvf fdt2.tar.gz -C ~/examples/fsl_course_data && tar zxvf tbss.tar.gz -C ~/examples/fsl_course_data; fi

nipype/interfaces/dipy/__init__.py

@@ -1,4 +1,5 @@
-from .tracks import TrackDensityMap
+from .tracks import StreamlineTractography, TrackDensityMap
 from .tensors import TensorMode, DTI
 from .preprocess import Resample, Denoise
+from .reconstruction import RESTORE, EstimateResponseSH, CSD
 from .simulate import SimulateMultiTensor

nipype/interfaces/dipy/base.py

@@ -0,0 +1,87 @@
+# -*- coding: utf-8 -*-
+""" Base interfaces for dipy """
+import os.path as op
+import numpy as np
+from nipype.interfaces.base import (traits, File, isdefined,
+                                    BaseInterface, BaseInterfaceInputSpec)
+from ... import logging
+
+IFLOGGER = logging.getLogger('interface')
+
+HAVE_DIPY = True
+try:
+    import dipy
+except ImportError:
+    HAVE_DIPY = False
+
+
+def no_dipy():
+    """ Check if dipy is available """
+    global HAVE_DIPY
+    return not HAVE_DIPY
+
+
+def dipy_version():
+    """ Check dipy version """
+    if no_dipy():
+        return None
+
+    return dipy.__version__
+
+
+class DipyBaseInterface(BaseInterface):
+
+    """
+    A base interface for py:mod:`dipy` computations
+    """
+    def __init__(self, **inputs):
+        if no_dipy():
+            IFLOGGER.error('dipy was not found')
+            # raise ImportError('dipy was not found')
+        super(DipyBaseInterface, self).__init__(**inputs)
+
+
+class DipyBaseInterfaceInputSpec(BaseInterfaceInputSpec):
+    in_file = File(exists=True, mandatory=True, desc=('input diffusion data'))
+    in_bval = File(exists=True, mandatory=True, desc=('input b-values table'))
+    in_bvec = File(exists=True, mandatory=True, desc=('input b-vectors table'))
+    b0_thres = traits.Int(700, usedefault=True, desc=('b0 threshold'))
+    out_prefix = traits.Str(desc=('output prefix for file names'))
+
+
+class DipyDiffusionInterface(DipyBaseInterface):
+
+    """
+    A base interface for py:mod:`dipy` computations
+    """
+    input_spec = DipyBaseInterfaceInputSpec
+
+    def _get_gradient_table(self):
+        bval = np.loadtxt(self.inputs.in_bval)
+        bvec = np.loadtxt(self.inputs.in_bvec).T
+        try:
+            from dipy.data import GradientTable
+            gtab = GradientTable(bvec)
+            gtab.bvals = bval
+        except NameError:
+            from dipy.core.gradients import gradient_table
+            gtab = gradient_table(bval, bvec)
+
+        gtab.b0_threshold = self.inputs.b0_thres
+        return gtab
+
+    def _gen_filename(self, name, ext=None):
+        fname, fext = op.splitext(op.basename(self.inputs.in_file))
+        if fext == '.gz':
+            fname, fext2 = op.splitext(fname)
+            fext = fext2 + fext
+
+        if not isdefined(self.inputs.out_prefix):
+            out_prefix = op.abspath(fname)
+        else:
+            out_prefix = self.inputs.out_prefix
+
+        if ext is None:
+            ext = fext
+
+        return out_prefix + '_' + name + ext

nipype/interfaces/dipy/preprocess.py

@@ -1,28 +1,20 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
-"""Change directory to provide relative paths for doctests
+"""
+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)
 """
 import os.path as op
-import warnings
-
 import nibabel as nb
 import numpy as np
 
-from ..base import (traits, TraitedSpec, BaseInterface, File, isdefined)
-from ...utils.filemanip import split_filename
-from ...utils.misc import package_check
+from ..base import (traits, TraitedSpec, File, isdefined)
+from .base import DipyBaseInterface
 from ... import logging
-iflogger = logging.getLogger('interface')
-
-have_dipy = True
-try:
-    package_check('dipy', version='0.6.0')
-except Exception as e:
-    have_dipy = False
+IFLOGGER = logging.getLogger('interface')
 
 
 class ResampleInputSpec(TraitedSpec):
@@ -33,15 +25,17 @@ class ResampleInputSpec(TraitedSpec):
                                   ' is set, then isotropic regridding will '
                                   'be performed, with spacing equal to the '
                                   'smallest current zoom.'))
-    interp = traits.Int(1, mandatory=True, usedefault=True, desc=('order of '
-                        'the interpolator (0 = nearest, 1 = linear, etc.'))
+    interp = traits.Int(
+        1, mandatory=True, usedefault=True,
+        desc=('order of the interpolator (0 = nearest, 1 = linear, etc.'))
 
 
 class ResampleOutputSpec(TraitedSpec):
     out_file = File(exists=True)
 
 
-class Resample(BaseInterface):
+class Resample(DipyBaseInterface):
+
     """
     An interface to reslicing diffusion datasets.
     See
@@ -69,7 +63,7 @@ def _run_interface(self, runtime):
         resample_proxy(self.inputs.in_file, order=order,
                        new_zooms=vox_size, out_file=out_file)
 
-        iflogger.info('Resliced image saved as {i}'.format(i=out_file))
+        IFLOGGER.info('Resliced image saved as {i}'.format(i=out_file))
         return runtime
 
     def _list_outputs(self):
@@ -92,17 +86,21 @@ class DenoiseInputSpec(TraitedSpec):
     noise_model = traits.Enum('rician', 'gaussian', mandatory=True,
                               usedefault=True,
                               desc=('noise distribution model'))
+    signal_mask = File(desc=('mask in which the mean signal '
+                             'will be computed'), exists=True)
     noise_mask = File(desc=('mask in which the standard deviation of noise '
                             'will be computed'), exists=True)
     patch_radius = traits.Int(1, desc='patch radius')
     block_radius = traits.Int(5, desc='block_radius')
+    snr = traits.Float(desc='manually set an SNR')
 
 
 class DenoiseOutputSpec(TraitedSpec):
     out_file = File(exists=True)
 
 
-class Denoise(BaseInterface):
+class Denoise(DipyBaseInterface):
+
     """
     An interface to denoising diffusion datasets [Coupe2008]_.
     See
@@ -140,16 +138,24 @@ def _run_interface(self, runtime):
         if isdefined(self.inputs.block_radius):
             settings['block_radius'] = self.inputs.block_radius
 
+        snr = None
+        if isdefined(self.inputs.snr):
+            snr = self.inputs.snr
+
+        signal_mask = None
+        if isdefined(self.inputs.signal_mask):
+            signal_mask = nb.load(self.inputs.signal_mask).get_data()
         noise_mask = None
-        if isdefined(self.inputs.in_mask):
+        if isdefined(self.inputs.noise_mask):
             noise_mask = nb.load(self.inputs.noise_mask).get_data()
 
-        _, s = nlmeans_proxy(self.inputs.in_file,
-                             settings,
-                             noise_mask=noise_mask,
+        _, s = nlmeans_proxy(self.inputs.in_file, settings,
+                             snr=snr,
+                             smask=signal_mask,
+                             nmask=noise_mask,
                              out_file=out_file)
-        iflogger.info(('Denoised image saved as {i}, estimated '
-                      'sigma={s}').format(i=out_file, s=s))
+        IFLOGGER.info(('Denoised image saved as {i}, estimated '
+                       'SNR={s}').format(i=out_file, s=str(s)))
         return runtime
 
     def _list_outputs(self):
@@ -169,7 +175,7 @@ def resample_proxy(in_file, order=3, new_zooms=None, out_file=None):
     """
     Performs regridding of an image to set isotropic voxel sizes using dipy.
     """
-    from dipy.align.aniso2iso import resample
+    from dipy.align.reslice import reslice
 
     if out_file is None:
         fname, fext = op.splitext(op.basename(in_file))
@@ -191,7 +197,7 @@ def resample_proxy(in_file, order=3, new_zooms=None, out_file=None):
     if np.all(im_zooms == new_zooms):
         return in_file
 
-    data2, affine2 = resample(data, affine, im_zooms, new_zooms, order=order)
+    data2, affine2 = reslice(data, affine, im_zooms, new_zooms, order=order)
     tmp_zooms = np.array(hdr.get_zooms())
     tmp_zooms[:3] = new_zooms[0]
     hdr.set_zooms(tuple(tmp_zooms))
@@ -203,12 +209,16 @@ def resample_proxy(in_file, order=3, new_zooms=None, out_file=None):
 
 
 def nlmeans_proxy(in_file, settings,
-                  noise_mask=None, out_file=None):
+                  snr=None,
+                  smask=None,
+                  nmask=None,
+                  out_file=None):
     """
     Uses non-local means to denoise 4D datasets
     """
-    package_check('dipy', version='0.8.0.dev')
     from dipy.denoise.nlmeans import nlmeans
+    from scipy.ndimage.morphology import binary_erosion
+    from scipy import ndimage
 
     if out_file is None:
         fname, fext = op.splitext(op.basename(in_file))
@@ -222,13 +232,69 @@ def nlmeans_proxy(in_file, settings,
     data = img.get_data()
     aff = img.affine
 
-    nmask = data[..., 0] > 80
-    if noise_mask is not None:
-        nmask = noise_mask > 0
-
-    sigma = np.std(data[nmask == 1])
-    den = nlmeans(data, sigma, **settings)
+    if data.ndim < 4:
+        data = data[..., np.newaxis]
+
+    data = np.nan_to_num(data)
+
+    if data.max() < 1.0e-4:
+        raise RuntimeError('There is no signal in the image')
+
+    df = 1.0
+    if data.max() < 1000.0:
+        df = 1000. / data.max()
+        data *= df
+
+    b0 = data[..., 0]
+
+    if smask is None:
+        smask = np.zeros_like(b0)
+        smask[b0 > np.percentile(b0, 85.)] = 1
+
+    smask = binary_erosion(
+        smask.astype(np.uint8), iterations=2).astype(np.uint8)
+
+    if nmask is None:
+        nmask = np.ones_like(b0, dtype=np.uint8)
+        bmask = settings['mask']
+        if bmask is None:
+            bmask = np.zeros_like(b0)
+            bmask[b0 > np.percentile(b0[b0 > 0], 10)] = 1
+            label_im, nb_labels = ndimage.label(bmask)
+            sizes = ndimage.sum(bmask, label_im, range(nb_labels + 1))
+            maxidx = np.argmax(sizes)
+            bmask = np.zeros_like(b0, dtype=np.uint8)
+            bmask[label_im == maxidx] = 1
+        nmask[bmask > 0] = 0
+    else:
+        nmask = np.squeeze(nmask)
+        nmask[nmask > 0.0] = 1
+        nmask[nmask < 1] = 0
+        nmask = nmask.astype(bool)
+
+    nmask = binary_erosion(nmask, iterations=1).astype(np.uint8)
+
+    den = np.zeros_like(data)
+
+    est_snr = True
+    if snr is not None:
+        snr = [snr] * data.shape[-1]
+        est_snr = False
+    else:
+        snr = []
+
+    for i in range(data.shape[-1]):
+        d = data[..., i]
+        if est_snr:
+            s = np.mean(d[smask > 0])
+            n = np.std(d[nmask > 0])
+            snr.append(s / n)
+
+        den[..., i] = nlmeans(d, snr[i], **settings)
+
+    den = np.squeeze(den)
+    den /= df
 
     nb.Nifti1Image(den.astype(hdr.get_data_dtype()), aff,
                    hdr).to_filename(out_file)
-    return out_file, sigma
+    return out_file, snr