File size: 3,341 Bytes
ee594f0 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 | # ANTsPy (https://github.com/ANTsX/ANTsPy)
import ants
def registration_function(fixed_image_path, moving_image_path, output_image_path):
"""
Registers a moving image to a fixed image using affine transformation and saves the registered image.
Parameters:
fixed_image_path (str): Path to the fixed image (reference image).
moving_image_path (str): Path to the moving image (image to be registered).
output_image_path (str): Path where the registered image will be saved.
"""
# Step 1: Load images using ANTs
fixed_image = ants.image_read(fixed_image_path) # Read the fixed (reference) image
moving_image = ants.image_read(moving_image_path) # Read the moving image
'''
ANTs registration function returns a dictionary containing:
- warpedmovout: The moving image registered to the fixed image
- warpedfixout: The fixed image registered to the moving image
- fwdtransforms: Transformation field from moving to fixed image
- invtransforms: Transformation field from fixed to moving image
Available transform types:
- Rigid: Rigid transformation (translation + rotation)
- Affine: Affine transformation (rigid + scaling)
- ElasticSyN: Elastic SyN transformation (affine + nonlinear with MI as the criterion)
- SyN: SyN transformation (affine + nonlinear with MI as the criterion)
- SyNCC: SyN transformation (affine + nonlinear with CC as the criterion)
'''
# Step 2: Perform image registration (using affine transformation)
registration_result = ants.registration(fixed=fixed_image, moving=moving_image, type_of_transform='Affine')
# Step 3: Apply the forward transformation to the moving image to get the registered image
registered_image = ants.apply_transforms(
fixed=fixed_image, # Fixed image (reference)
moving=moving_image, # Moving image (to be registered)
transformlist=registration_result['fwdtransforms'], # Transformation field
interpolator="linear" # Interpolation method for resampling
)
# Step 4: Ensure the registered image matches the fixed image's properties (direction, origin, spacing)
registered_image.set_direction(fixed_image.direction)
registered_image.set_origin(fixed_image.origin)
registered_image.set_spacing(fixed_image.spacing)
# Step 5: Save the registered image
ants.image_write(registered_image, output_image_path)
print("Registration complete. Registered image saved to:", output_image_path)
if __name__ == '__main__':
# Define the paths for fixed and moving images, and the output paths for the registered images
fixed_image_path = "./regdata/spgr_unstrip_lps_b.nii.gz"
moving_image_path1 = "./regdata/LGG-637/T1C.nii.gz"
output_image_path1 = "./regdata/LGG-637/registration_T1C.nii.gz"
moving_image_path2 = "./regdata/LGG-637/T2W.nii.gz"
output_image_path2 = "./regdata/LGG-637/registration_T2W.nii.gz"
# Perform registration for the first moving image (T1C)
registration_function(fixed_image_path, moving_image_path1, output_image_path1)
# Perform registration for the second moving image (T2w) using the registered T1C image as the fixed image
registration_function(output_image_path1, moving_image_path2, output_image_path2) |