def test_reconst_ivim():
with TemporaryDirectory() as out_dir:
bvals = np.array([
0., 10., 20., 30., 40., 60., 80., 100., 120., 140., 160., 180.,
200., 300., 400., 500., 600., 700., 800., 900., 1000.
])
N = len(bvals)
bvecs = generate_bvecs(N)
temp_bval_path = pjoin(out_dir, "temp.bval")
np.savetxt(temp_bval_path, bvals)
temp_bvec_path = pjoin(out_dir, "temp.bvec")
np.savetxt(temp_bvec_path, bvecs)
gtab = gradient_table(bvals, bvecs)
S0, f, D_star, D = 1000.0, 0.132, 0.00885, 0.000921
mevals = np.array(([D_star, D_star, D_star], [D, D, D]))
# This gives an isotropic signal.
data = multi_tensor(gtab,
mevals,
snr=None,
S0=S0,
fractions=[f * 100, 100 * (1 - f)])
# Single voxel data
data_single = data[0]
temp_affine = np.eye(4)
data_multi = np.zeros((2, 2, 1, len(gtab.bvals)), dtype=int)
data_multi[0, 0, 0] = data_multi[0, 1, 0] = data_multi[
1, 0, 0] = data_multi[1, 1, 0] = data_single
data_path = pjoin(out_dir, 'tmp_data.nii.gz')
save_nifti(data_path, data_multi, temp_affine)
mask = np.ones_like(data_multi[..., 0], dtype=np.uint8)
mask_path = pjoin(out_dir, 'tmp_mask.nii.gz')
save_nifti(mask_path, mask, temp_affine)
ivim_flow = ReconstIvimFlow()
args = [data_path, temp_bval_path, temp_bvec_path, mask_path]
ivim_flow.run(*args, out_dir=out_dir)
S0_path = ivim_flow.last_generated_outputs['out_S0_predicted']
S0_data = load_nifti_data(S0_path)
assert_equal(S0_data.shape, data_multi.shape[:-1])
f_path = ivim_flow.last_generated_outputs['out_perfusion_fraction']
f_data = load_nifti_data(f_path)
assert_equal(f_data.shape, data_multi.shape[:-1])
D_star_path = ivim_flow.last_generated_outputs['out_D_star']
D_star_data = load_nifti_data(D_star_path)
assert_equal(D_star_data.shape, data_multi.shape[:-1])
D_path = ivim_flow.last_generated_outputs['out_D']
D_data = load_nifti_data(D_path)
assert_equal(D_data.shape, data_multi.shape[:-1])
def test_motion_correction():
data_path, fbvals_path, fbvecs_path = get_fnames('small_64D')
volume = load_nifti_data(data_path)
with TemporaryDirectory() as out_dir:
# Use an abbreviated data-set:
img = nib.load(data_path)
data = img.get_fdata()[..., :10]
nib.save(nib.Nifti1Image(data, img.affine),
os.path.join(out_dir, 'data.nii.gz'))
# Save a subset:
bvals = np.loadtxt(fbvals_path)
bvecs = np.loadtxt(fbvecs_path)
np.savetxt(os.path.join(out_dir, 'bvals.txt'), bvals[:10])
np.savetxt(os.path.join(out_dir, 'bvecs.txt'), bvecs[:10])
motion_correction_flow = MotionCorrectionFlow()
motion_correction_flow._force_overwrite = True
motion_correction_flow.run(os.path.join(out_dir, 'data.nii.gz'),
os.path.join(out_dir, 'bvals.txt'),
os.path.join(out_dir, 'bvecs.txt'),
out_dir=out_dir)
out_path = motion_correction_flow.last_generated_outputs['out_moved']
corrected = load_nifti_data(out_path)
npt.assert_equal(corrected.shape, data.shape)
npt.assert_equal(corrected.min(), data.min())
npt.assert_equal(corrected.max(), data.max())
def _cmc_sc(self):
from dipy.tracking.stopping_criterion import CmcStoppingCriterion
f_pve_csf, f_pve_gm, f_pve_wm = load_pve_files(self.subj_folder,
self.pve_file_name)[:3]
pve_csf_data = load_nifti_data(f_pve_csf)
pve_gm_data = load_nifti_data(f_pve_gm)
pve_wm_data = load_nifti_data(f_pve_wm)
cmc_criterion = CmcStoppingCriterion.from_pve(
pve_wm_data,
pve_gm_data,
pve_csf_data,
step_size=self.parameters_dict['step_size'],
average_voxel_size=self.parameters_dict['voxel_size'])
self.classifier = cmc_criterion
def test_response_from_mask():
fdata, fbvals, fbvecs = get_fnames('small_64D')
bvals, bvecs = read_bvals_bvecs(fbvals, fbvecs)
data = load_nifti_data(fdata)
gtab = gradient_table(bvals, bvecs)
ten = TensorModel(gtab)
tenfit = ten.fit(data)
FA = fractional_anisotropy(tenfit.evals)
FA[np.isnan(FA)] = 0
radius = 3
for fa_thr in np.arange(0, 1, 0.1):
response_auto, ratio_auto, nvoxels = auto_response(
gtab,
data,
roi_center=None,
roi_radius=radius,
fa_thr=fa_thr,
return_number_of_voxels=True)
ci, cj, ck = np.array(data.shape[:3]) // 2
mask = np.zeros(data.shape[:3])
mask[ci - radius:ci + radius, cj - radius:cj + radius,
ck - radius:ck + radius] = 1
mask[FA <= fa_thr] = 0
response_mask, ratio_mask = response_from_mask(gtab, data, mask)
assert_equal(int(np.sum(mask)), nvoxels)
assert_array_almost_equal(response_mask[0], response_auto[0])
assert_almost_equal(response_mask[1], response_auto[1])
assert_almost_equal(ratio_mask, ratio_auto)
def test_reslice():
with TemporaryDirectory() as out_dir:
data_path, _, _ = get_fnames('small_25')
volume = load_nifti_data(data_path)
reslice_flow = ResliceFlow()
reslice_flow.run(data_path, [1.5, 1.5, 1.5], out_dir=out_dir)
out_path = reslice_flow.last_generated_outputs['out_resliced']
resliced = load_nifti_data(out_path)
npt.assert_equal(resliced.shape[0] > volume.shape[0], True)
npt.assert_equal(resliced.shape[1] > volume.shape[1], True)
npt.assert_equal(resliced.shape[2] > volume.shape[2], True)
npt.assert_equal(resliced.shape[-1], volume.shape[-1])
def test_dti_xval():
data = load_nifti_data(fdata)
gtab = gt.gradient_table(fbval, fbvec)
dm = dti.TensorModel(gtab, 'LS')
# The data has 102 directions, so will not divide neatly into 10 bits
npt.assert_raises(ValueError, xval.kfold_xval, dm, data, 10)
# In simulation with no noise, COD should be perfect:
psphere = dpd.get_sphere('symmetric362')
bvecs = np.concatenate(([[0, 0, 0]], psphere.vertices))
bvals = np.zeros(len(bvecs)) + 1000
bvals[0] = 0
gtab = gt.gradient_table(bvals, bvecs)
mevals = np.array(([0.0015, 0.0003, 0.0001], [0.0015, 0.0003, 0.0003]))
mevecs = [
np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1]]),
np.array([[0, 0, 1], [0, 1, 0], [1, 0, 0]])
]
S = sims.single_tensor(gtab, 100, mevals[0], mevecs[0], snr=None)
dm = dti.TensorModel(gtab, 'LS')
kf_xval = xval.kfold_xval(dm, S, 2)
cod = xval.coeff_of_determination(S, kf_xval)
npt.assert_array_almost_equal(cod, np.ones(kf_xval.shape[:-1]) * 100)
# Test with 2D data for use of a mask
S = np.array([[S, S], [S, S]])
mask = np.ones(S.shape[:-1], dtype=bool)
mask[1, 1] = 0
kf_xval = xval.kfold_xval(dm, S, 2, mask=mask)
cod2d = xval.coeff_of_determination(S, kf_xval)
npt.assert_array_almost_equal(np.round(cod2d[0, 0]), cod)
def test_exponential_iso():
fdata, fbvals, fbvecs = dpd.get_fnames()
data_dti = load_nifti_data(fdata)
gtab_dti = grad.gradient_table(fbvals, fbvecs)
data_multi, gtab_multi = dpd.dsi_deconv_voxels()
for data, gtab in zip([data_dti, data_multi], [gtab_dti, gtab_multi]):
sfmodel = sfm.SparseFascicleModel(
gtab, isotropic=sfm.ExponentialIsotropicModel)
sffit1 = sfmodel.fit(data[0, 0, 0])
sphere = dpd.get_sphere()
sffit1.odf(sphere)
sffit1.predict(gtab)
SNR = 1000
S0 = 100
mevals = np.array(([0.0015, 0.0005, 0.0005], [0.0015, 0.0005, 0.0005]))
angles = [(0, 0), (60, 0)]
S, sticks = sims.multi_tensor(gtab,
mevals,
S0,
angles=angles,
fractions=[50, 50],
snr=SNR)
sffit = sfmodel.fit(S)
pred = sffit.predict()
npt.assert_(xval.coeff_of_determination(pred, S) > 96)
def test_mppca_flow():
with TemporaryDirectory() as out_dir:
S0 = 100 + 2 * np.random.standard_normal((22, 23, 30, 20))
data_path = os.path.join(out_dir, "random_noise.nii.gz")
save_nifti(data_path, S0, np.eye(4))
mppca_flow = MPPCAFlow()
mppca_flow.run(data_path, out_dir=out_dir)
assert_true(os.path.isfile(
mppca_flow.last_generated_outputs['out_denoised']))
assert_false(os.path.isfile(
mppca_flow.last_generated_outputs['out_sigma']))
mppca_flow._force_overwrite = True
mppca_flow.run(data_path, return_sigma=True, pca_method='svd',
out_dir=out_dir)
assert_true(os.path.isfile(
mppca_flow.last_generated_outputs['out_denoised']))
assert_true(os.path.isfile(
mppca_flow.last_generated_outputs['out_sigma']))
denoised_path = mppca_flow.last_generated_outputs['out_denoised']
denoised_data = load_nifti_data(denoised_path)
assert_greater(denoised_data.min(), S0.min())
assert_less(denoised_data.max(), S0.max())
npt.assert_equal(np.round(denoised_data.mean()), 100)
def test_FiberFit():
data_file, bval_file, bvec_file = dpd.get_fnames('small_64D')
data = load_nifti_data(data_file)
bvals, bvecs = read_bvals_bvecs(bval_file, bvec_file)
gtab = grad.gradient_table(bvals, bvecs)
FM = life.FiberModel(gtab)
evals = [0.0015, 0.0005, 0.0005]
streamline = [[[1, 2, 3], [4, 5, 3], [5, 6, 3], [6, 7, 3]],
[[1, 2, 3], [4, 5, 3], [5, 6, 3]]]
fiber_matrix, vox_coords = FM.setup(streamline, np.eye(4), evals)
w = np.array([0.5, 0.5])
sig = opt.spdot(fiber_matrix, w) + 1.0 # Add some isotropic stuff
S0 = data[..., gtab.b0s_mask]
this_data = np.zeros((10, 10, 10, 64))
this_data[vox_coords[:, 0], vox_coords[:, 1], vox_coords[:, 2]] =\
(sig.reshape((4, 64)) *
S0[vox_coords[:, 0], vox_coords[:, 1], vox_coords[:, 2]])
# Grab some realistic S0 values:
this_data = np.concatenate([data[..., gtab.b0s_mask], this_data], -1)
fit = FM.fit(this_data, streamline, np.eye(4))
npt.assert_almost_equal(fit.predict()[1], fit.data[1], decimal=-1)
# Predict with an input GradientTable
npt.assert_almost_equal(fit.predict(gtab)[1], fit.data[1], decimal=-1)
npt.assert_almost_equal(
this_data[vox_coords[:, 0], vox_coords[:, 1], vox_coords[:, 2]],
fit.data)
def test_sfm_stick():
fdata, fbvals, fbvecs = dpd.get_fnames()
data = load_nifti_data(fdata)
gtab = grad.gradient_table(fbvals, fbvecs)
sfmodel = sfm.SparseFascicleModel(gtab,
solver='NNLS',
response=[0.001, 0, 0])
sffit1 = sfmodel.fit(data[0, 0, 0])
sphere = dpd.get_sphere()
sffit1.odf(sphere)
sffit1.predict(gtab)
SNR = 1000
S0 = 100
mevals = np.array(([0.001, 0, 0], [0.001, 0, 0]))
angles = [(0, 0), (60, 0)]
S, sticks = sims.multi_tensor(gtab,
mevals,
S0,
angles=angles,
fractions=[50, 50],
snr=SNR)
sfmodel = sfm.SparseFascicleModel(gtab,
solver='NNLS',
response=[0.001, 0, 0])
sffit = sfmodel.fit(S)
pred = sffit.predict()
npt.assert_(xval.coeff_of_determination(pred, S) > 96)
def test_sfm():
fdata, fbvals, fbvecs = dpd.get_fnames()
data = load_nifti_data(fdata)
gtab = grad.gradient_table(fbvals, fbvecs)
for iso in [sfm.ExponentialIsotropicModel, None]:
sfmodel = sfm.SparseFascicleModel(gtab, isotropic=iso)
sffit1 = sfmodel.fit(data[0, 0, 0])
sphere = dpd.get_sphere()
odf1 = sffit1.odf(sphere)
pred1 = sffit1.predict(gtab)
mask = np.ones(data.shape[:-1])
sffit2 = sfmodel.fit(data, mask)
pred2 = sffit2.predict(gtab)
odf2 = sffit2.odf(sphere)
sffit3 = sfmodel.fit(data)
pred3 = sffit3.predict(gtab)
odf3 = sffit3.odf(sphere)
npt.assert_almost_equal(pred3, pred2, decimal=2)
npt.assert_almost_equal(pred3[0, 0, 0], pred1, decimal=2)
npt.assert_almost_equal(odf3[0, 0, 0], odf1, decimal=2)
npt.assert_almost_equal(odf3[0, 0, 0], odf2[0, 0, 0], decimal=2)
# Fit zeros and you will get back zeros
npt.assert_almost_equal(
sfmodel.fit(np.zeros(data[0, 0, 0].shape)).beta,
np.zeros(sfmodel.design_matrix[0].shape[-1]))
def do_segmentation(self,src_path,dst_path):
"""
This function is used to segment the input image into a segmented image based on grey matter, white matter and csf. These are the features used to
detect the stage of AD
Variables & its function:
nclass => value is initialized as 3 inorder to divide to 3 classes-GM,WM,CSF
beta => The smoothness factor of segmentation
t0 => Stores the time before segmentation
hmrf => Create an instance of class TissueClassifierHMRF
t1 => Store the time after segmentation
total_time => Calculate the total time taken for segmentation
brain => Reconstructing the segmented brain image as a 3D file which is then saved to the destination
"""
#To get only the image data array
t1 = load_nifti_data(src_path)
print('t1.shape (%d, %d, %d)' % t1.shape)
#To load the entire nifti file
t2 = nib.load(src_path)
nclass = 3
beta = 0.1
t0 = time.time()
hmrf = TissueClassifierHMRF()
#Perform segmentation
initial_segmentation, final_segmentation, PVE = hmrf.classify(t1, nclass, beta)
t1 = time.time()
total_time = t1-t0
print('Total time:' + str(total_time))
print(final_segmentation.shape)
brain = nib.Nifti1Image(final_segmentation,t2.affine)
print('Segmentation performed successfully')
nib.save(brain, os.path.join(dst_path))
def test_median_otsu():
fname = get_fnames('S0_10')
data = load_nifti_data(fname)
data = np.squeeze(data.astype('f8'))
dummy_mask = data > data.mean()
data_masked, mask = median_otsu(data, median_radius=3, numpass=2,
autocrop=False, vol_idx=None,
dilate=None)
assert_equal(mask.sum() < dummy_mask.sum(), True)
data2 = np.zeros(data.shape + (2,))
data2[..., 0] = data
data2[..., 1] = data
data2_masked, mask2 = median_otsu(data2, median_radius=3, numpass=2,
autocrop=False, vol_idx=[0, 1],
dilate=None)
assert_almost_equal(mask.sum(), mask2.sum())
_, mask3 = median_otsu(data2, median_radius=3, numpass=2,
autocrop=False, vol_idx=[0, 1],
dilate=1)
assert_equal(mask2.sum() < mask3.sum(), True)
_, mask4 = median_otsu(data2, median_radius=3, numpass=2,
autocrop=False, vol_idx=[0, 1],
dilate=2)
assert_equal(mask3.sum() < mask4.sum(), True)
# For 4D volumes, can't call without vol_idx input:
assert_raises(ValueError, median_otsu, data2)
def eval_method(name=None, method=None, track_path=None, data_path=None):
if track_path == None:
track_path = './Result/Track/tractogram_' + method + '_' + name + '.trk'
if data_path == None:
data_path = './data/DWI/' + name + '/'
if not op.exists(track_path):
print('no tracking')
return 0
else:
from dipy.io.gradients import read_bvals_bvecs
from dipy.io.image import load_nifti_data, load_nifti
from dipy.core.gradients import gradient_table
data, affine, hardi_img = load_nifti(data_path + 'norm.nii.gz',
return_img=True)
print(data.shape)
labels = load_nifti_data(data_path + 'seg.nii.gz')
# t1_data = load_nifti_data('./data/tanenci_20170601/b0.nii.gz')
bvals, bvecs = read_bvals_bvecs(data_path + 'DWI.bval',
data_path + 'DWI.bvec')
gtab = gradient_table(bvals, bvecs)
# Read the candidates from file in voxel space:
candidate_sl_sft = load_trk(track_path, 'same', bbox_valid_check=False)
candidate_sl_sft.to_vox()
candidate_sl = candidate_sl_sft.streamlines
print('loading finished, begin weighting')
fiber_model = life.FiberModel(gtab)
inv_affine = np.linalg.inv(hardi_img.affine)
fiber_fit = fiber_model.fit(data,
reduct(candidate_sl, data[:, :, :, 0]),
affine=np.eye(4))
print('weighting finished, begin prediction')
beta_baseline = np.zeros(fiber_fit.beta.shape[0])
pred_weighted = np.reshape(
opt.spdot(fiber_fit.life_matrix, beta_baseline),
(fiber_fit.vox_coords.shape[0], np.sum(~gtab.b0s_mask)))
model_predict = fiber_fit.predict()
model_error = model_predict - fiber_fit.data
model_rmse = np.sqrt(np.mean(model_error[:, 10:]**2, -1))
#print('model_rmse:', model_rmse.shape)
vol_model = np.zeros(data.shape[:3]) * np.nan
vol_model[fiber_fit.vox_coords[:, 0], fiber_fit.vox_coords[:, 1],
fiber_fit.vox_coords[:, 2]] = model_rmse
#print('error:', np.sum(vol_model) / model_rmse.shape[0])
return np.sum(model_rmse) / model_rmse.shape[0], vol_model, affine
def create_cmc_classifier(folder_name):
from dipy.tracking.stopping_criterion import CmcStoppingCriterion
f_pve_csf, f_pve_gm, f_pve_wm = load_pve_files(folder_name)
pve_csf_data = load_nifti_data(f_pve_csf)
pve_gm_data = load_nifti_data(f_pve_gm)
pve_wm_data, _, voxel_size = load_nifti(f_pve_wm, return_voxsize=True)
voxel_size = np.average(voxel_size[1:4])
step_size = 0.2
cmc_criterion = CmcStoppingCriterion.from_pve(
pve_wm_data,
pve_gm_data,
pve_csf_data,
step_size=step_size,
average_voxel_size=voxel_size)
return cmc_criterion, step_size
def read_cenir_multib(bvals=None):
"""Read CENIR multi b-value data.
Parameters
----------
bvals : list or int
The b-values to read from file (200, 400, 1000, 2000, 3000).
Returns
-------
gtab : a GradientTable class instance
img : nibabel.Nifti1Image
"""
files, folder = fetch_cenir_multib(with_raw=False)
if bvals is None:
bvals = [200, 400, 1000, 2000, 3000]
if isinstance(bvals, int):
bvals = [bvals]
file_dict = {
200: {
'DWI': pjoin(folder, '4D_dwi_eddycor_B200.nii.gz'),
'bvals': pjoin(folder, 'dwi_bvals_B200'),
'bvecs': pjoin(folder, 'dwi_bvecs_B200')
},
400: {
'DWI': pjoin(folder, '4D_dwieddycor_B400.nii.gz'),
'bvals': pjoin(folder, 'bvals_B400'),
'bvecs': pjoin(folder, 'bvecs_B400')
},
1000: {
'DWI': pjoin(folder, '4D_dwieddycor_B1000.nii.gz'),
'bvals': pjoin(folder, 'bvals_B1000'),
'bvecs': pjoin(folder, 'bvecs_B1000')
},
2000: {
'DWI': pjoin(folder, '4D_dwieddycor_B2000.nii.gz'),
'bvals': pjoin(folder, 'bvals_B2000'),
'bvecs': pjoin(folder, 'bvecs_B2000')
},
3000: {
'DWI': pjoin(folder, '4D_dwieddycor_B3000.nii.gz'),
'bvals': pjoin(folder, 'bvals_B3000'),
'bvecs': pjoin(folder, 'bvecs_B3000')
}
}
data = []
bval_list = []
bvec_list = []
for bval in bvals:
data.append(load_nifti_data(file_dict[bval]['DWI']))
bval_list.extend(np.loadtxt(file_dict[bval]['bvals']))
bvec_list.append(np.loadtxt(file_dict[bval]['bvecs']))
# All affines are the same, so grab the last one:
aff = nib.load(file_dict[bval]['DWI']).affine
return (nib.Nifti1Image(np.concatenate(data, -1), aff),
gradient_table(bval_list, np.concatenate(bvec_list, -1)))
def test_sfm_background():
fdata, fbvals, fbvecs = dpd.get_fnames()
data = load_nifti_data(fdata)
gtab = grad.gradient_table(fbvals, fbvecs)
to_fit = data[0, 0, 0]
to_fit[gtab.b0s_mask] = 0
sfmodel = sfm.SparseFascicleModel(gtab, solver='NNLS')
sffit = sfmodel.fit(to_fit)
npt.assert_equal(sffit.beta, np.zeros_like(sffit.beta))
def test_odf_with_zeros():
fdata, fbval, fbvec = get_fnames('small_25')
gtab = grad.gradient_table(fbval, fbvec)
data = load_nifti_data(fdata)
dm = dti.TensorModel(gtab)
df = dm.fit(data)
df.evals[0, 0, 0] = np.array([0, 0, 0])
sphere = create_unit_sphere(4)
odf = df.odf(sphere)
npt.assert_equal(odf[0, 0, 0], np.zeros(sphere.vertices.shape[0]))
def reconst_mmri_core(flow, lap, pos):
with TemporaryDirectory() as out_dir:
data_path, bval_path, bvec_path = get_fnames('small_25')
volume = load_nifti_data(data_path)
mmri_flow = flow()
mmri_flow.run(data_files=data_path,
bvals_files=bval_path,
bvecs_files=bvec_path,
small_delta=0.0129,
big_delta=0.0218,
laplacian=lap,
positivity=pos,
out_dir=out_dir)
for out_name in [
'out_rtop', 'out_lapnorm', 'out_msd', 'out_qiv', 'out_rtap',
'out_rtpp', 'out_ng', 'out_parng', 'out_perng'
]:
out_path = mmri_flow.last_generated_outputs[out_name]
out_data = load_nifti_data(out_path)
npt.assert_equal(out_data.shape, volume.shape[:-1])
bvals, bvecs = read_bvals_bvecs(bval_path, bvec_path)
bvals[0] = 5.
bvecs = generate_bvecs(len(bvals))
tmp_bval_path = pjoin(out_dir, "tmp.bval")
tmp_bvec_path = pjoin(out_dir, "tmp.bvec")
np.savetxt(tmp_bval_path, bvals)
np.savetxt(tmp_bvec_path, bvecs.T)
mmri_flow._force_overwrite = True
with npt.assert_raises(BaseException):
npt.assert_warns(UserWarning,
mmri_flow.run,
data_path,
tmp_bval_path,
tmp_bvec_path,
small_delta=0.0129,
big_delta=0.0218,
laplacian=lap,
positivity=pos,
out_dir=out_dir)
def get_data(name=None,
data_path=None,
pft=False,
norm='norm.nii.gz',
bval='DWI.bval',
bvec='DWI.bvec',
seg='seg.nii.gz',
FA=None,
mask='norm_mask.nii.gz'):
if name != None:
root = './data/DWI/' + name
else:
root = data_path
DWI = root + '/' + norm
bvals_path = root + '/' + bval
bvecs_path = root + '/' + bvec
mask_path = root + '/' + mask
data, affine, img = load_nifti(DWI, return_img=True)
if seg == None:
labels = None
else:
seg_path = root + '/' + seg
labels = load_nifti_data(seg_path)
bvals, bvecs = read_bvals_bvecs(bvals_path, bvecs_path)
gtab = gradient_table(bvals, bvecs)
head_mask = load_nifti_data(mask_path)
if not pft:
return data, affine, img, labels, gtab, head_mask
else:
f_pve_csf = root + './pve_2.nii.gz'
f_pve_gm = root + './pve_0.nii.gz'
f_pve_wm = root + './pve_1.nii.gz'
pve_csf_data = load_nifti_data(f_pve_csf)
pve_gm_data = load_nifti_data(f_pve_gm)
pve_wm_data, _, voxel_size = load_nifti(f_pve_wm, return_voxsize=True)
return data, affine, img, labels, gtab, head_mask, \
pve_csf_data, pve_gm_data, pve_wm_data, voxel_size
def test_min_signal_alone():
fdata, fbvals, fbvecs = get_fnames()
data = load_nifti_data(fdata)
gtab = grad.gradient_table(fbvals, fbvecs)
idx = tuple(np.array(np.where(data == np.min(data)))[:-1, 0])
ten_model = dti.TensorModel(gtab)
fit_alone = ten_model.fit(data[idx])
fit_together = ten_model.fit(data)
npt.assert_array_almost_equal(fit_together.model_params[idx],
fit_alone.model_params, decimal=12)
def bench_csdeconv(center=(50, 40, 40), width=12):
img, gtab, labels_img = read_stanford_labels()
data = load_nifti_data(img)
labels = load_nifti_data(labels_img)
shape = labels.shape
mask = np.in1d(labels, [1, 2])
mask.shape = shape
a, b, c = center
hw = width // 2
idx = (slice(a - hw, a + hw), slice(b - hw, b + hw), slice(c - hw, c + hw))
data_small = data[idx].copy()
mask_small = mask[idx].copy()
voxels = mask_small.sum()
cmd = "model.fit(data_small, mask_small)"
print("== Benchmarking CSD fit on %d voxels ==" % voxels)
msg = "SH order - %d, gradient directons - %d :: %g sec"
# Basic case
sh_order = 8
ConstrainedSphericalDeconvModel(gtab, None, sh_order=sh_order)
time = npt.measure(cmd)
print(msg % (sh_order, num_grad(gtab), time))
# Smaller data set
# data_small = data_small[..., :75].copy()
gtab = GradientTable(gtab.gradients[:75])
ConstrainedSphericalDeconvModel(gtab, None, sh_order=sh_order)
time = npt.measure(cmd)
print(msg % (sh_order, num_grad(gtab), time))
# Super resolution
sh_order = 12
ConstrainedSphericalDeconvModel(gtab, None, sh_order=sh_order)
time = npt.measure(cmd)
print(msg % (sh_order, num_grad(gtab), time))
def test_median_otsu():
fname = get_fnames('S0_10')
data = load_nifti_data(fname)
data = cp.asarray(np.squeeze(data.astype('f8')))
dummy_mask = data > data.mean()
data_masked, mask = median_otsu(
data,
median_radius=3,
numpass=2,
autocrop=False,
vol_idx=None,
dilate=None,
)
assert mask.sum() < dummy_mask.sum()
data2 = cp.zeros(data.shape + (2, ))
data2[..., 0] = data
data2[..., 1] = data
data2_masked, mask2 = median_otsu(
data2,
median_radius=3,
numpass=2,
autocrop=False,
vol_idx=[0, 1],
dilate=None,
)
assert mask.sum() == mask2.sum()
_, mask3 = median_otsu(
data2,
median_radius=3,
numpass=2,
autocrop=False,
vol_idx=[0, 1],
dilate=1,
)
assert mask2.sum() < mask3.sum()
_, mask4 = median_otsu(
data2,
median_radius=3,
numpass=2,
autocrop=False,
vol_idx=[0, 1],
dilate=2,
)
assert mask3.sum() < mask4.sum()
# For 4D volumes, can't call without vol_idx input:
with pytest.raises(ValueError):
median_otsu(data2)
def test_csd_xval():
# First, let's see that it works with some data:
data = load_nifti_data(fdata)[1:3, 1:3, 1:3] # Make it *small*
gtab = gt.gradient_table(fbval, fbvec)
S0 = np.mean(data[..., gtab.b0s_mask])
response = ([0.0015, 0.0003, 0.0001], S0)
# In simulation, it should work rather well (high COD):
psphere = dpd.get_sphere('symmetric362')
bvecs = np.concatenate(([[0, 0, 0]], psphere.vertices))
bvals = np.zeros(len(bvecs)) + 1000
bvals[0] = 0
gtab = gt.gradient_table(bvals, bvecs)
mevals = np.array(([0.0015, 0.0003, 0.0001], [0.0015, 0.0003, 0.0003]))
mevecs = [
np.array([[1, 0, 0], [0, 1, 0], [0, 0, 1]]),
np.array([[0, 0, 1], [0, 1, 0], [1, 0, 0]])
]
S0 = 100
S = sims.single_tensor(gtab, S0, mevals[0], mevecs[0], snr=None)
with warnings.catch_warnings():
warnings.filterwarnings("ignore",
message=descoteaux07_legacy_msg,
category=PendingDeprecationWarning)
sm = csd.ConstrainedSphericalDeconvModel(gtab, response)
np.random.seed(12345)
response = ([0.0015, 0.0003, 0.0001], S0)
with warnings.catch_warnings():
warnings.filterwarnings("ignore",
message=descoteaux07_legacy_msg,
category=PendingDeprecationWarning)
kf_xval = xval.kfold_xval(sm, S, 2, response, sh_order=2)
# Because of the regularization, COD is not going to be perfect here:
cod = xval.coeff_of_determination(S, kf_xval)
# We'll just test for regressions:
csd_cod = 97 # pre-computed by hand for this random seed
# We're going to be really lenient here:
npt.assert_array_almost_equal(np.round(cod), csd_cod)
# Test for sD data with more than one voxel for use of a mask:
S = np.array([[S, S], [S, S]])
mask = np.ones(S.shape[:-1], dtype=bool)
mask[1, 1] = 0
with warnings.catch_warnings():
warnings.filterwarnings("ignore",
message=descoteaux07_legacy_msg,
category=PendingDeprecationWarning)
kf_xval = xval.kfold_xval(sm, S, 2, response, sh_order=2, mask=mask)
cod = xval.coeff_of_determination(S, kf_xval)
npt.assert_array_almost_equal(np.round(cod[0]), csd_cod)
def test_median_otsu_flow():
with TemporaryDirectory() as out_dir:
data_path, _, _ = get_fnames('small_25')
volume = load_nifti_data(data_path)
save_masked = True
median_radius = 3
numpass = 3
autocrop = False
vol_idx = [0]
dilate = 0
mo_flow = MedianOtsuFlow()
mo_flow.run(data_path,
out_dir=out_dir,
save_masked=save_masked,
median_radius=median_radius,
numpass=numpass,
autocrop=autocrop,
vol_idx=vol_idx,
dilate=dilate)
mask_name = mo_flow.last_generated_outputs['out_mask']
masked_name = mo_flow.last_generated_outputs['out_masked']
masked, mask = median_otsu(volume,
vol_idx=vol_idx,
median_radius=median_radius,
numpass=numpass,
autocrop=autocrop,
dilate=dilate)
result_mask_data = load_nifti_data(join(out_dir, mask_name))
npt.assert_array_equal(result_mask_data.astype(np.uint8), mask)
result_masked = nib.load(join(out_dir, masked_name))
result_masked_data = np.asanyarray(result_masked.dataobj)
npt.assert_array_equal(np.round(result_masked_data), masked)
def reconst_flow_core(flow, extra_args=[], extra_kwargs={}):
with TemporaryDirectory() as out_dir:
data_path, bval_path, bvec_path = get_fnames('small_25')
volume, affine = load_nifti(data_path)
mask = np.ones_like(volume[:, :, :, 0], dtype=np.uint8)
mask_path = join(out_dir, 'tmp_mask.nii.gz')
save_nifti(mask_path, mask, affine)
dti_flow = flow()
args = [data_path, bval_path, bvec_path, mask_path]
args.extend(extra_args)
kwargs = dict(out_dir=out_dir)
kwargs.update(extra_kwargs)
dti_flow.run(*args, **kwargs)
fa_path = dti_flow.last_generated_outputs['out_fa']
fa_data = load_nifti_data(fa_path)
assert_equal(fa_data.shape, volume.shape[:-1])
tensor_path = dti_flow.last_generated_outputs['out_tensor']
tensor_data = load_nifti_data(tensor_path)
# Per default, tensor data is 5D, with six tensor elements on the last
# dimension, except if nifti_tensor is set to False:
if extra_kwargs.get('nifti_tensor', True):
assert_equal(tensor_data.shape[-1], 6)
assert_equal(tensor_data.shape[:-2], volume.shape[:-1])
else:
assert_equal(tensor_data.shape[-1], 6)
assert_equal(tensor_data.shape[:-1], volume.shape[:-1])
for out_name in ['out_ga', 'out_md', 'out_ad', 'out_rd', 'out_mode']:
out_path = dti_flow.last_generated_outputs[out_name]
out_data = load_nifti_data(out_path)
assert_equal(out_data.shape, volume.shape[:-1])
rgb_path = dti_flow.last_generated_outputs['out_rgb']
rgb_data = load_nifti_data(rgb_path)
assert_equal(rgb_data.shape[-1], 3)
assert_equal(rgb_data.shape[:-1], volume.shape[:-1])
evecs_path = dti_flow.last_generated_outputs['out_evec']
evecs_data = load_nifti_data(evecs_path)
assert_equal(evecs_data.shape[-2:], tuple((3, 3)))
assert_equal(evecs_data.shape[:-2], volume.shape[:-1])
evals_path = dti_flow.last_generated_outputs['out_eval']
evals_data = load_nifti_data(evals_path)
assert_equal(evals_data.shape[-1], 3)
assert_equal(evals_data.shape[:-1], volume.shape[:-1])
def test_auto_response():
fdata, fbvals, fbvecs = get_fnames('small_64D')
bvals, bvecs = read_bvals_bvecs(fbvals, fbvecs)
data = load_nifti_data(fdata)
gtab = gradient_table(bvals, bvecs)
radius = 3
def test_fa_superior(FA, fa_thr):
return FA > fa_thr
def test_fa_inferior(FA, fa_thr):
return FA < fa_thr
predefined_functions = [fa_superior, fa_inferior]
defined_functions = [test_fa_superior, test_fa_inferior]
for fa_thr in np.arange(0.1, 1, 0.1):
for predefined, defined in \
zip(predefined_functions, defined_functions):
response_predefined, ratio_predefined, nvoxels_predefined = \
auto_response(gtab,
data,
roi_center=None,
roi_radius=radius,
fa_callable=predefined,
fa_thr=fa_thr,
return_number_of_voxels=True)
response_defined, ratio_defined, nvoxels_defined = \
auto_response(gtab,
data,
roi_center=None,
roi_radius=radius,
fa_callable=defined,
fa_thr=fa_thr,
return_number_of_voxels=True)
assert_equal(nvoxels_predefined, nvoxels_defined)
assert_array_almost_equal(response_predefined[0],
response_defined[0])
assert_almost_equal(response_predefined[1], response_defined[1])
assert_almost_equal(ratio_predefined, ratio_defined)
def test_no_predict():
# Test that if you try to do this with a model that doesn't have a `predict`
# method, you get something reasonable.
class NoPredictModel(base.ReconstModel):
def __init__(self, gtab):
base.ReconstModel.__init__(self, gtab)
def fit(self, data, mask=None):
return NoPredictFit(self, data, mask=mask)
class NoPredictFit(base.ReconstFit):
def __init__(self, model, data, mask=None):
base.ReconstFit.__init__(self, model, data)
gtab = gt.gradient_table(fbval, fbvec)
my_model = NoPredictModel(gtab)
data = load_nifti_data(fdata)[1:3, 1:3, 1:3] # Whatever
npt.assert_raises(ValueError, xval.kfold_xval, my_model, data, 2)
from dipy.reconst.csdeconv import (ConstrainedSphericalDeconvModel,
auto_response)
from dipy.tracking import utils
from dipy.tracking.local_tracking import LocalTracking
from dipy.tracking.streamline import Streamlines
from dipy.tracking.stopping_criterion import ThresholdStoppingCriterion
from dipy.viz import window, actor, colormap, has_fury
# Enables/disables interactive visualization
interactive = False
hardi_fname, hardi_bval_fname, hardi_bvec_fname = get_fnames('stanford_hardi')
label_fname = get_fnames('stanford_labels')
data, affine, hardi_img = load_nifti(hardi_fname, return_img=True)
labels = load_nifti_data(label_fname)
bvals, bvecs = read_bvals_bvecs(hardi_bval_fname, hardi_bvec_fname)
gtab = gradient_table(bvals, bvecs)
seed_mask = (labels == 2)
white_matter = (labels == 1) | (labels == 2)
seeds = utils.seeds_from_mask(seed_mask, affine, density=1)
response, ratio = auto_response(gtab, data, roi_radius=10, fa_thr=0.7)
csd_model = ConstrainedSphericalDeconvModel(gtab, response, sh_order=6)
csd_fit = csd_model.fit(data, mask=white_matter)
"""
We use the GFA of the CSA model to build a stopping criterion.
"""
from dipy.reconst.shm import CsaOdfModel
def load_TOM(fn):
data = load_nifti_data(fn)
return data