def objective_function(optimise_array, static_image, dynamic_path, dvf_path,
weighted_normalise, dynamic_data_magnitude):
static_image.fill(
np.reshape(optimise_array,
static_image.as_array().astype(np.double).shape))
objective_value = 0.0
for i in range(len(dynamic_path)):
dynamic_image = reg.NiftiImageData(dynamic_path[i])
dvf_image = reg.NiftiImageData3DDeformation(dvf_path[i])
resampler = reg.NiftyResample()
resampler.set_reference_image(static_image)
resampler.set_floating_image(dynamic_image)
resampler.add_transformation(dvf_image)
resampler.set_interpolation_type_to_cubic_spline()
objective_value = objective_value + (np.nansum(
np.square(dynamic_image.as_array().astype(np.double) -
((np.nansum(dynamic_image.as_array().astype(np.double),
dtype=np.double) / dynamic_data_magnitude) *
warp_image_forward(resampler, static_image)),
dtype=np.double),
dtype=np.double) * weighted_normalise[i])
print("Objective function value: {0}".format(str(objective_value)))
return objective_value
def get_resamplers(static_image, dynamic_array, dvf_array, output_path):
resamplers = []
static_image_path = "{0}/temp_static.nii".format(output_path)
dynamic_array_path = "{0}/temp_dynamic.nii".format(output_path)
dvf_array_path = "{0}/temp_dvf.nii".format(output_path)
for j in range(len(dynamic_array)):
resampler = reg.NiftyResample()
static_image.write(static_image_path)
dynamic_array[j].write(dynamic_array_path)
dvf_array[j].write(dvf_array_path)
temp_static = reg.NiftiImageData(static_image_path)
temp_dynamic = reg.NiftiImageData(dynamic_array_path)
temp_dvf = reg.NiftiImageData3DDeformation(dvf_array_path)
resampler.set_reference_image(temp_static)
resampler.set_floating_image(temp_dynamic)
resampler.add_transformation(temp_dvf)
resampler.set_interpolation_type_to_linear()
resamplers.append(resampler)
return resamplers
def output_input(static_image, dynamic_path, dvf_path, output_path):
static_image.write("{0}/static_image.nii".format(output_path))
for i in range(len(dynamic_path)):
dynamic_image = reg.NiftiImageData(dynamic_path[i])
dvf_image = reg.NiftiImageData3DDeformation(dvf_path[i])
dynamic_image.write("{0}/dynamic_image_{1}.nii".format(
output_path, str(i)))
dvf_image.write("{0}/dvf_image_{1}.nii".format(output_path, str(i)))
return True
def test_for_adj(static_image, dvf_array, output_path):
static_image_path = "{0}/temp_static.nii".format(output_path)
dvf_array_path = "{0}/temp_dvf.nii".format(output_path)
for i in range(len(dvf_array)):
static_image.write(static_image_path)
dvf_array[i].write(dvf_array_path)
temp_static = reg.NiftiImageData(static_image_path)
temp_dvf = reg.NiftiImageData3DDeformation(dvf_array_path)
resampler = reg.NiftyResample()
resampler.set_reference_image(temp_static)
resampler.set_floating_image(temp_static)
resampler.add_transformation(temp_dvf)
resampler.set_interpolation_type_to_linear()
warp = warp_image_forward(resampler, temp_static)
warped_image = static_image.clone()
warped_image.fill(warp)
warped_image.write("{0}/warp_forward_{1}.nii".format(
output_path, str(i)))
difference = temp_static.as_array().astype(np.double) - warp
difference_image = temp_static.clone()
difference_image.fill(difference)
difference_image.write("{0}/warp_forward_difference_{1}.nii".format(
output_path, str(i)))
warp = warp_image_adjoint(resampler, temp_static)
warped_image = temp_static.clone()
warped_image.fill(warp)
warped_image.write("{0}/warp_adjoint_{1}.nii".format(
output_path, str(i)))
difference = temp_static.as_array().astype(np.double) - warp
difference_image = temp_static.clone()
difference_image.fill(difference)
difference_image.write("{0}/warp_adjoint_difference_{1}.nii".format(
output_path, str(i)))
return True
def gradient_function(optimise_array, static_image, dynamic_path, dvf_path,
weighted_normalise, dynamic_data_magnitude):
static_image.fill(
np.reshape(optimise_array,
static_image.as_array().astype(np.double).shape))
gradient_value = static_image.clone()
gradient_value.fill(0.0)
adjoint_image = static_image.clone()
for i in range(len(dynamic_path)):
dynamic_image = reg.NiftiImageData(dynamic_path[i])
dvf_image = reg.NiftiImageData3DDeformation(dvf_path[i])
resampler = reg.NiftyResample()
resampler.set_reference_image(static_image)
resampler.set_floating_image(dynamic_image)
resampler.add_transformation(dvf_image)
resampler.set_interpolation_type_to_cubic_spline()
adjoint_image.fill((
(np.nansum(dynamic_image.as_array().astype(np.double),
dtype=np.double) / dynamic_data_magnitude) *
warp_image_forward(resampler, static_image)) -
dynamic_image.as_array().astype(np.double))
gradient_value.fill((gradient_value.as_array().astype(np.double) +
(warp_image_adjoint(resampler, adjoint_image) *
weighted_normalise[i])))
# gradient_value.write("{0}/gradient.nii".format(output_path))
print(
"Max gradient value: {0}, Mean gradient value: {1}, Gradient norm: {2}"
.format(
str(np.amax(gradient_value.as_array().astype(np.double))),
str(
np.nanmean(
np.abs(gradient_value.as_array().astype(np.double),
dtype=np.double))),
str(np.linalg.norm(gradient_value.as_array().astype(np.double)))))
return np.ravel(gradient_value.as_array().astype(np.double)).astype(
np.double)
def test_for_adj(static_image, dvf_path, output_path):
for i in range(len(dvf_path)):
dvf_image = reg.NiftiImageData3DDeformation(dvf_path[i])
resampler = reg.NiftyResample()
resampler.set_reference_image(static_image)
resampler.set_floating_image(static_image)
resampler.add_transformation(dvf_image)
resampler.set_interpolation_type_to_cubic_spline()
warp = warp_image_forward(resampler, static_image)
warped_image = static_image.clone()
warped_image.fill(warp)
warped_image.write("{0}/warp_forward_{1}.nii".format(
output_path, str(i)))
difference = static_image.as_array().astype(np.double) - warp
difference_image = static_image.clone()
difference_image.fill(difference)
difference_image.write("{0}/warp_forward_difference_{1}.nii".format(
output_path, str(i)))
warp = warp_image_adjoint(resampler, static_image)
warped_image = static_image.clone()
warped_image.fill(warp)
warped_image.write("{0}/warp_adjoint_{1}.nii".format(
output_path, str(i)))
difference = static_image.as_array().astype(np.double) - warp
difference_image = static_image.clone()
difference_image.fill(difference)
difference_image.write("{0}/warp_adjoint_difference_{1}.nii".format(
output_path, str(i)))
return True
def resample_attn_image(image):
"""Resample the attenuation image."""
if trans_type == 'tm':
transformation = reg.AffineTransformation(trans)
elif trans_type == 'disp':
transformation = reg.NiftiImageData3DDisplacement(trans)
elif trans_type == 'def':
transformation = reg.NiftiImageData3DDeformation(trans)
else:
raise ValueError("Unknown transformation type.")
resampler = reg.NiftyResample()
resampler.set_reference_image(image)
resampler.set_floating_image(image)
resampler.set_interpolation_type_to_linear()
resampler.set_padding_value(0.0)
resampler.add_transformation(transformation)
return resampler.forward(image)
def back_warp(static_path, dvf_path, output_path):
if not os.path.exists(output_path):
os.makedirs(output_path, mode=0o770)
for i in range(len(dvf_path)):
static_image = reg.NiftiImageData(static_path)
dvf_image = reg.NiftiImageData3DDeformation(dvf_path[i])
resampler = reg.NiftyResample()
resampler.set_reference_image(static_image)
resampler.set_floating_image(static_image)
resampler.add_transformation(dvf_image)
resampler.set_interpolation_type_to_cubic_spline()
warped_static_image = warp_image_forward(resampler, static_image)
static_image.fill(warped_static_image)
static_image.write("{0}/back_warped_{1}.nii".format(
output_path, str(i)))
return True
def main():
# file paths to data
input_data_path = parser.parser(sys.argv[1], "data_path:=")
data_split = parser.parser(sys.argv[1], "data_split:=")
input_dvf_path = parser.parser(sys.argv[1], "dvf_path:=")
dvf_split = parser.parser(sys.argv[1], "dvf_split:=")
output_path = parser.parser(sys.argv[1], "output_path:=")
do_op_test = parser.parser(sys.argv[1], "do_op_test:=")
do_reg = parser.parser(sys.argv[1], "do_reg:=")
do_test_for_adj = parser.parser(sys.argv[1], "do_test_for_adj:=")
for i in range(len(input_data_path)):
if not os.path.exists(output_path[i]):
os.makedirs(output_path[i], mode=0o770)
new_dvf_path = "{0}/new_dvfs/".format(output_path[i])
if not os.path.exists(new_dvf_path):
os.makedirs(new_dvf_path, mode=0o770)
# get static and dynamic paths
dynamic_path = get_data_path(input_data_path[i], data_split[i])
# load dynamic objects
dynamic_array = []
for j in range(len(dynamic_path)):
dynamic_array.append(reg.NiftiImageData(dynamic_path[j]))
static_path = "{0}/static_path.nii".format(output_path[i])
# load static objects
static_image = reg.NiftiImageData(dynamic_path[0])
for j in range(1, len(dynamic_path)):
static_image.fill(static_image.as_array().astype(np.double) +
dynamic_array[j].as_array().astype(np.double))
static_image.write(static_path)
if bool(distutils.util.strtobool(do_op_test[i])):
op_test(static_image, output_path[i])
# if do reg the calc dvf if not load
if bool(distutils.util.strtobool(do_reg[i])):
dvf_path = register_data(static_path, dynamic_path, output_path[i])
else:
dvf_path = get_dvf_path(input_dvf_path[i], dvf_split[i])
# fix dvf header and load dvf objects
dvf_path = edit_header(dvf_path, new_dvf_path)
dvf_array = []
for j in range(len(dvf_path)):
dvf_array.append(reg.NiftiImageData3DDeformation(dvf_path[j]))
# create object to get forward and adj
resamplers = get_resamplers(static_image, dynamic_array, dvf_array,
output_path[i])
# test for adj
if bool(distutils.util.strtobool(do_test_for_adj[i])):
test_for_adj(static_image, dvf_array, output_path[i])
output_input(static_image, dynamic_array, dvf_array,
output_path[i])
# initial static image
initial_static_image = static_image.clone()
# array to optimise
optimise_array = static_image.as_array().astype(np.double)
# array bounds
bounds = []
for j in range(len(np.ravel(optimise_array.copy()))):
bounds.append((-np.inf, np.inf))
# optimise
optimise_array = np.reshape(
scipy.optimize.minimize(objective_function,
np.ravel(optimise_array).astype(np.double),
args=(resamplers, dynamic_array,
static_image, output_path[i]),
method="L-BFGS-B",
jac=gradient_function,
bounds=bounds,
tol=0.0000000001,
options={
"disp": True
}).x, optimise_array.shape)
# output
static_image.fill(optimise_array)
static_image.write("{0}/optimiser_output_{1}.nii".format(
output_path[i], str(i)))
difference = static_image.as_array().astype(
np.double) - initial_static_image.as_array().astype(np.double)
difference_image = initial_static_image.clone()
difference_image.fill(difference)
static_image.write("{0}/optimiser_output_difference_{1}.nii".format(
output_path[i], str(i)))
def main():
###########################################################################
# Parse input files
###########################################################################
if trans_pattern is None:
raise AssertionError("--trans missing")
if sino_pattern is None:
raise AssertionError("--sino missing")
trans_files = sorted(glob(trans_pattern))
sino_files = sorted(glob(sino_pattern))
attn_files = sorted(glob(attn_pattern))
rand_files = sorted(glob(rand_pattern))
num_ms = len(sino_files)
# Check some sinograms found
if num_ms == 0:
raise AssertionError("No sinograms found!")
# Should have as many trans as sinos
if num_ms != len(trans_files):
raise AssertionError("#trans should match #sinos. "
"#sinos = " + str(num_ms) + ", #trans = " +
str(len(trans_files)))
# If any rand, check num == num_ms
if len(rand_files) > 0 and len(rand_files) != num_ms:
raise AssertionError("#rand should match #sinos. "
"#sinos = " + str(num_ms) + ", #rand = " +
str(len(rand_files)))
# For attn, there should be 0, 1 or num_ms images
if len(attn_files) > 1 and len(attn_files) != num_ms:
raise AssertionError("#attn should be 0, 1 or #sinos")
###########################################################################
# Read input
###########################################################################
if trans_type == "tm":
trans = [reg.AffineTransformation(file) for file in trans_files]
elif trans_type == "disp":
trans = [
reg.NiftiImageData3DDisplacement(file) for file in trans_files
]
elif trans_type == "def":
trans = [reg.NiftiImageData3DDeformation(file) for file in trans_files]
else:
raise error("Unknown transformation type")
sinos_raw = [pet.AcquisitionData(file) for file in sino_files]
attns = [pet.ImageData(file) for file in attn_files]
rands = [pet.AcquisitionData(file) for file in rand_files]
# Loop over all sinograms
sinos = [0] * num_ms
for ind in range(num_ms):
# If any sinograms contain negative values
# (shouldn't be the case), set them to 0
sino_arr = sinos_raw[ind].as_array()
if (sino_arr < 0).any():
print("Input sinogram " + str(ind) +
" contains -ve elements. Setting to 0...")
sinos[ind] = sinos_raw[ind].clone()
sino_arr[sino_arr < 0] = 0
sinos[ind].fill(sino_arr)
else:
sinos[ind] = sinos_raw[ind]
# If rebinning is desired
segs_to_combine = 1
if args['--numSegsToCombine']:
segs_to_combine = int(args['--numSegsToCombine'])
views_to_combine = 1
if args['--numViewsToCombine']:
views_to_combine = int(args['--numViewsToCombine'])
if segs_to_combine * views_to_combine > 1:
sinos[ind] = sinos[ind].rebin(segs_to_combine, views_to_combine)
# only print first time
if ind == 0:
print(f"Rebinned sino dimensions: {sinos[ind].dimensions()}")
###########################################################################
# Initialise recon image
###########################################################################
if initial_estimate:
image = pet.ImageData(initial_estimate)
else:
# Create image based on ProjData
image = sinos[0].create_uniform_image(0.0, (nxny, nxny))
# If using GPU, need to make sure that image is right size.
if use_gpu:
dim = (127, 320, 320)
spacing = (2.03125, 2.08626, 2.08626)
# elif non-default spacing desired
elif args['--dxdy']:
dim = image.dimensions()
dxdy = float(args['--dxdy'])
spacing = (image.voxel_sizes()[0], dxdy, dxdy)
if use_gpu or args['--dxdy']:
image.initialise(dim=dim, vsize=spacing)
image.fill(0.0)
###########################################################################
# Set up resamplers
###########################################################################
resamplers = [get_resampler(image, trans=tran) for tran in trans]
###########################################################################
# Resample attenuation images (if necessary)
###########################################################################
resampled_attns = None
if len(attns) > 0:
resampled_attns = [0] * num_ms
# if using GPU, dimensions of attn and recon images have to match
ref = image if use_gpu else None
for i in range(len(attns)):
# if we only have 1 attn image, then we need to resample into
# space of each gate. However, if we have num_ms attn images, then
# assume they are already in the correct position, so use None as
# transformation.
tran = trans[i] if len(attns) == 1 else None
# If only 1 attn image, then resample that. If we have num_ms attn
# images, then use each attn image of each frame.
attn = attns[0] if len(attns) == 1 else attns[i]
resam = get_resampler(attn, ref=ref, trans=tran)
resampled_attns[i] = resam.forward(attn)
###########################################################################
# Set up acquisition models
###########################################################################
print("Setting up acquisition models...")
if not use_gpu:
acq_models = num_ms * [pet.AcquisitionModelUsingRayTracingMatrix()]
else:
acq_models = num_ms * [pet.AcquisitionModelUsingNiftyPET()]
for acq_model in acq_models:
acq_model.set_use_truncation(True)
acq_model.set_cuda_verbosity(verbosity)
# If present, create ASM from ECAT8 normalisation data
asm_norm = None
if norm_file:
asm_norm = pet.AcquisitionSensitivityModel(norm_file)
# Loop over each motion state
for ind in range(num_ms):
# Create attn ASM if necessary
asm_attn = None
if resampled_attns:
asm_attn = get_asm_attn(sinos[ind], resampled_attns[i],
acq_models[ind])
# Get ASM dependent on attn and/or norm
asm = None
if asm_norm and asm_attn:
if ind == 0:
print("ASM contains norm and attenuation...")
asm = pet.AcquisitionSensitivityModel(asm_norm, asm_attn)
elif asm_norm:
if ind == 0:
print("ASM contains norm...")
asm = asm_norm
elif asm_attn:
if ind == 0:
print("ASM contains attenuation...")
asm = asm_attn
if asm:
acq_models[ind].set_acquisition_sensitivity(asm)
if len(rands) > 0:
acq_models[ind].set_background_term(rands[ind])
# Set up
acq_models[ind].set_up(sinos[ind], image)
###########################################################################
# Set up reconstructor
###########################################################################
print("Setting up reconstructor...")
# Create composition operators containing acquisition models and resamplers
C = [
CompositionOperator(am, res, preallocate=True)
for am, res in zip(*(acq_models, resamplers))
]
# Configure the PDHG algorithm
if args['--normK'] and not args['--onlyNormK']:
normK = float(args['--normK'])
else:
kl = [KullbackLeibler(b=sino, eta=(sino * 0 + 1e-5)) for sino in sinos]
f = BlockFunction(*kl)
K = BlockOperator(*C)
# Calculate normK
print("Calculating norm of the block operator...")
normK = K.norm(iterations=10)
print("Norm of the BlockOperator ", normK)
if args['--onlyNormK']:
exit(0)
# Optionally rescale sinograms and BlockOperator using normK
scale_factor = 1. / normK if args['--normaliseDataAndBlock'] else 1.0
kl = [
KullbackLeibler(b=sino * scale_factor, eta=(sino * 0 + 1e-5))
for sino in sinos
]
f = BlockFunction(*kl)
K = BlockOperator(*C) * scale_factor
# If preconditioned
if precond:
def get_nonzero_recip(data):
"""Get the reciprocal of a datacontainer. Voxels where input == 0
will have their reciprocal set to 1 (instead of infinity)"""
inv_np = data.as_array()
inv_np[inv_np == 0] = 1
inv_np = 1. / inv_np
data.fill(inv_np)
tau = K.adjoint(K.range_geometry().allocate(1))
get_nonzero_recip(tau)
tmp_sigma = K.direct(K.domain_geometry().allocate(1))
sigma = 0. * tmp_sigma
get_nonzero_recip(sigma[0])
def precond_proximal(self, x, tau, out=None):
"""Modify proximal method to work with preconditioned tau"""
pars = {
'algorithm':
FGP_TV,
'input':
np.asarray(x.as_array() / tau.as_array(), dtype=np.float32),
'regularization_parameter':
self.lambdaReg,
'number_of_iterations':
self.iterationsTV,
'tolerance_constant':
self.tolerance,
'methodTV':
self.methodTV,
'nonneg':
self.nonnegativity,
'printingOut':
self.printing
}
res, info = regularisers.FGP_TV(pars['input'],
pars['regularization_parameter'],
pars['number_of_iterations'],
pars['tolerance_constant'],
pars['methodTV'], pars['nonneg'],
self.device)
if out is not None:
out.fill(res)
else:
out = x.copy()
out.fill(res)
out *= tau
return out
FGP_TV.proximal = precond_proximal
print("Will run proximal with preconditioned tau...")
# If not preconditioned
else:
sigma = float(args['--sigma'])
# If we need to calculate default tau
if args['--tau']:
tau = float(args['--tau'])
else:
tau = 1 / (sigma * normK**2)
if regularisation == 'none':
G = IndicatorBox(lower=0)
elif regularisation == 'FGP_TV':
r_iterations = float(args['--reg_iters'])
r_tolerance = 1e-7
r_iso = 0
r_nonneg = 1
r_printing = 0
device = 'gpu' if use_gpu else 'cpu'
G = FGP_TV(r_alpha, r_iterations, r_tolerance, r_iso, r_nonneg,
r_printing, device)
else:
raise error("Unknown regularisation")
if precond:
def PDHG_new_update(self):
"""Modify the PDHG update to allow preconditioning"""
# save previous iteration
self.x_old.fill(self.x)
self.y_old.fill(self.y)
# Gradient ascent for the dual variable
self.operator.direct(self.xbar, out=self.y_tmp)
self.y_tmp *= self.sigma
self.y_tmp += self.y_old
self.f.proximal_conjugate(self.y_tmp, self.sigma, out=self.y)
# Gradient descent for the primal variable
self.operator.adjoint(self.y, out=self.x_tmp)
self.x_tmp *= -1 * self.tau
self.x_tmp += self.x_old
self.g.proximal(self.x_tmp, self.tau, out=self.x)
# Update
self.x.subtract(self.x_old, out=self.xbar)
self.xbar *= self.theta
self.xbar += self.x
PDHG.update = PDHG_new_update
# Get filename
outp_file = outp_prefix
if descriptive_fname:
if len(attn_files) > 0:
outp_file += "_wAC"
if norm_file:
outp_file += "_wNorm"
if use_gpu:
outp_file += "_wGPU"
outp_file += "_Reg-" + regularisation
if regularisation == 'FGP_TV':
outp_file += "-alpha" + str(r_alpha)
outp_file += "-riters" + str(r_iterations)
if args['--normK']:
outp_file += '_userNormK' + str(normK)
else:
outp_file += '_calcNormK' + str(normK)
if args['--normaliseDataAndBlock']:
outp_file += '_wDataScale'
else:
outp_file += '_noDataScale'
if not precond:
outp_file += "_sigma" + str(sigma)
outp_file += "_tau" + str(tau)
else:
outp_file += "_wPrecond"
outp_file += "_nGates" + str(len(sino_files))
if resamplers is None:
outp_file += "_noMotion"
pdhg = PDHG(f=f,
g=G,
operator=K,
sigma=sigma,
tau=tau,
max_iteration=num_iters,
update_objective_interval=update_obj_fn_interval,
x_init=image,
log_file=outp_file + ".log")
def callback_save(iteration, objective_value, solution):
"""Callback function to save images"""
if (iteration + 1) % save_interval == 0:
out = solution if not nifti else reg.NiftiImageData(solution)
out.write(outp_file + "_iters" + str(iteration + 1))
pdhg.run(iterations=num_iters,
callback=callback_save,
verbose=True,
very_verbose=True)
if visualisations:
# show reconstructed image
out = pdhg.get_output()
out_arr = out.as_array()
z = out_arr.shape[0] // 2
show_2D_array('Reconstructed image', out.as_array()[z, :, :])
pylab.show()