def evaluate_ncpdnet(multicoil=False,
dcomp=False,
normalize_image=False,
n_iter=10,
n_filters=32,
n_primal=5,
non_linearity='relu',
**eval_kwargs):
run_params = {
'n_primal': n_primal,
'multicoil': multicoil,
'activation': non_linearity,
'n_iter': n_iter,
'n_filters': n_filters,
'im_size': IM_SIZE,
'dcomp': dcomp,
'normalize_image': normalize_image,
}
model = NCPDNet(**run_params)
return evaluate_nc(
model,
multicoil=multicoil,
dcomp=dcomp,
**eval_kwargs,
)
def evaluate_ncpdnet(multicoil=False,
three_d=False,
dcomp=False,
normalize_image=False,
n_iter=10,
n_filters=32,
n_primal=5,
non_linearity='relu',
refine_smaps=True,
**eval_kwargs):
if three_d:
image_size = VOLUME_SIZE
else:
image_size = IM_SIZE
run_params = {
'n_primal': n_primal,
'multicoil': multicoil,
'three_d': three_d,
'activation': non_linearity,
'n_iter': n_iter,
'n_filters': n_filters,
'im_size': image_size,
'dcomp': dcomp,
'normalize_image': normalize_image,
'refine_smaps': refine_smaps,
}
model = NCPDNet(**run_params)
return evaluate_nc(
model,
multicoil=multicoil,
dcomp=dcomp,
three_d=three_d,
**eval_kwargs,
)
def train_ncpdnet(
multicoil=False,
three_d=False,
dcomp=False,
normalize_image=False,
n_iter=10,
n_filters=32,
n_primal=5,
non_linearity='relu',
refine_smaps=True,
**train_kwargs,
):
if three_d:
image_size = VOLUME_SIZE
else:
image_size = IM_SIZE
run_params = {
'n_primal': n_primal,
'multicoil': multicoil,
'three_d': three_d,
'activation': non_linearity,
'n_iter': n_iter,
'n_filters': n_filters,
'im_size': image_size,
'dcomp': dcomp,
'normalize_image': normalize_image,
'refine_smaps': refine_smaps,
'fastmri': not three_d,
}
if multicoil:
ncpdnet_type = 'ncpdnet_sense_'
elif three_d:
ncpdnet_type = 'ncpdnet_3d_'
else:
ncpdnet_type = 'ncpdnet_singlecoil_'
additional_info = ''
if n_iter != 10:
additional_info += f'_i{n_iter}'
if non_linearity != 'relu':
additional_info += f'_{non_linearity}'
if multicoil and refine_smaps:
additional_info += '_rfs'
run_id = f'{ncpdnet_type}_{additional_info}'
model = NCPDNet(**run_params)
return train_ncnet_block(
model,
n_iter=n_iter,
run_id=run_id,
multicoil=multicoil,
dcomp=dcomp,
three_d=three_d,
**train_kwargs,
)
def test_ncpdnet_init_and_call(ktraj):
model = NCPDNet(n_iter=3, n_primal=3, n_filters=8)
image_shape = (640, 372)
nspokes = 15
traj = ktraj(image_shape, nspokes)
spokelength = image_shape[-1] * 2
kspace_shape = spokelength * nspokes
model([
tf.zeros([1, 1, kspace_shape, 1], dtype=tf.complex64),
traj,
(tf.constant([image_shape[-1]]), ),
])
def ncpdnet_qualitative_validation(multicoil=False,
three_d=False,
dcomp=False,
normalize_image=False,
n_iter=10,
n_filters=32,
n_primal=5,
non_linearity='relu',
refine_smaps=True,
brain=False,
**eval_kwargs):
if three_d:
image_size = VOLUME_SIZE
else:
image_size = IM_SIZE
run_params = {
'n_primal': n_primal,
'multicoil': multicoil,
'three_d': three_d,
'activation': non_linearity,
'n_iter': n_iter,
'n_filters': n_filters,
'im_size': image_size,
'dcomp': dcomp,
'normalize_image': normalize_image,
'refine_smaps': refine_smaps,
'output_shape_spec': brain,
'fastmri': not three_d,
}
model = NCPDNet(**run_params)
name = 'pdnet'
if dcomp:
name += '-dcomp'
if normalize_image:
name += '-norm'
return ncnet_qualitative_validation(
model,
name,
multicoil=multicoil,
dcomp=dcomp,
three_d=three_d,
brain=brain,
**eval_kwargs,
)
def test_ncpdnet_init_and_call_3d(dcomp, volume_shape):
model = NCPDNet(
n_iter=1,
n_primal=2,
n_filters=2,
multicoil=False,
im_size=volume_shape,
three_d=True,
dcomp=dcomp,
fastmri=False,
)
af = 16
traj = get_stacks_of_radial_trajectory(volume_shape, af=af)
spokelength = volume_shape[-2]
nspokes = volume_shape[-1] // af
nstacks = volume_shape[0]
kspace_shape = nspokes * spokelength * nstacks
extra_args = (tf.constant([volume_shape]), )
if dcomp:
nufft_ob = KbNufftModule(
im_size=volume_shape,
grid_size=None,
norm='ortho',
)
interpob = nufft_ob._extract_nufft_interpob()
nufftob_forw = kbnufft_forward(interpob)
nufftob_back = kbnufft_adjoint(interpob)
dcomp = calculate_radial_dcomp_tf(
interpob,
nufftob_forw,
nufftob_back,
traj[0],
stacks=True,
)
dcomp = tf.ones([1, tf.shape(dcomp)[0]],
dtype=dcomp.dtype) * dcomp[None, :]
extra_args += (dcomp, )
res = model([
tf.zeros([1, 1, kspace_shape, 1], dtype=tf.complex64),
traj,
extra_args,
])
assert res.shape[1:4] == volume_shape