def single(self, kspace, fname):
kspace = common.to_tensor(kspace)
# Apply mask
seed = None if not self.use_seed else tuple(map(ord, fname))
masked_kspace, mask = common.apply_mask(kspace, self.mask_func, seed)
# Inverse Fourier Transform to get zero filled solution
image = common.ifft2(masked_kspace)
# Crop input image
image = common.complex_center_crop(image,
(self.resolution, self.resolution))
# Absolute value
image = common.complex_abs(image)
# Apply Root-Sum-of-Squares if multicoil data
if self.which_challenge == 'multicoil':
image = common.root_sum_of_squares(image)
# Normalize input
image, mean, std = common.normalize_instance(image, eps=1e-11)
image = image.clamp(-6, 6)
return image, mean, std
def __call__(self, kspace, target, attrs, fname, slice):
"""
Args:
kspace (numpy.array): Input k-space of shape (num_coils, rows, cols, 2) for multi-coil
data or (rows, cols, 2) for single coil data.
target (numpy.array): Target image
attrs (dict): Acquisition related information stored in the HDF5 object.
fname (str): File name
slice (int): Serial number of the slice.
Returns:
(tuple): tuple containing:
image (torch.Tensor): Zero-filled input image.
target (torch.Tensor): Target image converted to a torch Tensor.
mean (float): Mean value used for normalization.
std (float): Standard deviation value used for normalization.
norm (float): L2 norm of the entire volume.
"""
kspace = common.to_tensor(kspace)
# Apply mask
seed = None if not self.use_seed else tuple(map(ord, fname))
masked_kspace, mask = common.apply_mask(kspace, self.mask_func, seed)
# Inverse Fourier Transform to get zero filled solution
image = common.ifft2(masked_kspace)
# Crop input image
image = common.complex_center_crop(image,
(self.resolution, self.resolution))
# Absolute value
image = common.complex_abs(image)
# Apply Root-Sum-of-Squares if multicoil data
if self.which_challenge == 'multicoil':
image = common.root_sum_of_squares(image)
# Normalize input
image, mean, std = common.normalize_instance(image, eps=1e-11)
image = image.clamp(-6, 6)
target = common.to_tensor(target)
# Normalize target
target = common.normalize(target, mean, std, eps=1e-11)
target = target.clamp(-6, 6)
# Apply random crop
if self.crop:
image, target = common.random_crop(image, target, self.crop_size)
return image, target, mean, std, attrs['norm'].astype(np.float32)
def to_img(kspace, target, which_challenge, mask_func, resolution):
kspace = common.to_tensor(kspace)
seed = tuple(map(ord, '233'))
masked_kspace = kspace
if mask_func is not None:
masked_kspace, mask = common.apply_mask(kspace, mask_func, seed)
# Inverse Fourier Transform to get zero filled solution
image = common.ifft2(masked_kspace)
# Crop input image
image = common.complex_center_crop(image, (resolution, resolution))
# Absolute value
image = common.complex_abs(image)
# Apply Root-Sum-of-Squares if multicoil data
if which_challenge == 'multicoil':
image = common.root_sum_of_squares(image)
image, mean, std = common.normalize_instance(image, eps=1e-11)
image = image.clamp(-6, 6)
target = common.to_tensor(target)
image, target = common.random_crop(image, target, 196)
# Normalize target
target = common.normalize(target, mean, std, eps=1e-11)
target = target.clamp(-6, 6)
return image.numpy(), target.numpy()
def __call__(self, kspace, target, attrs, fname, slice):
"""
Args:
kspace (numpy.Array): k-space measurements
target (numpy.Array): Target image
attrs (dict): Acquisition related information stored in the HDF5 object
fname (pathlib.Path): Path to the input file
slice (int): Serial number of the slice
Returns:
(tuple): tuple containing:
image (torch.Tensor): Normalized zero-filled input image
mean (float): Mean of the zero-filled image
std (float): Standard deviation of the zero-filled image
fname (pathlib.Path): Path to the input file
slice (int): Serial number of the slice
"""
kspace = common.to_tensor(kspace)
if self.mask_func is not None:
seed = tuple(map(ord, fname))
masked_kspace, _ = common.apply_mask(kspace, self.mask_func, seed)
else:
masked_kspace = kspace
# Inverse Fourier Transform to get zero filled solution
image = common.ifft2(masked_kspace)
# Crop input image
image = common.complex_center_crop(image,
(self.resolution, self.resolution))
# Absolute value
image = common.complex_abs(image)
# Apply Root-Sum-of-Squares if multicoil data
if self.which_challenge == 'multicoil':
image = common.root_sum_of_squares(image)
# Normalize input
image, mean, std = common.normalize_instance(image)
image = image.clamp(-6, 6)
return image, mean, std, fname, slice