mdreg 0.4.4

Model-driven image registration

Model-driven motion correction for medical imaging

• Documentation: https://qib-sheffield.github.io/mdreg/

• Source code: https://github.com/QIB-Sheffield/mdreg

Note: mdreg is under construction. At this stage, the API may still change and features may be deprecated without warning.

Installation

pip install mdreg

Typical usage

import mdreg

# Get some test data (variable flip angle MRI)
data = mdreg.fetch('VFA')

# Configure the signal model fit
fit_image = {
    'func': mdreg.spgr_vfa_lin,     # VFA signal model
    'FA': data['FA'],               # Flip angle in degress
    'progress_bar': True,           # Show a progress bar
}

# Configure the coregistration method
fit_coreg = {
    'package': 'elastix',
    'spacing': data['spacing'],
}

# Perform the motion correction
coreg, defo, fit, pars = mdreg.fit(
    data['array'],                # Signal data to correct
    fit_image = fit_image,        # Signal model fit
    fit_coreg = fit_coreg,        # Coregistration
    verbose = 2,                  # Show progress update
)

# Visualize the results
anim = mdreg.animation(
    coreg,
    title = 'Motion corrected VFA',
    'interval' : 500,                   # Time between animation frames in ms
    'vmin' : 0,                         # Minimum value of the colorbar
    'vmax' : np.percentile(coreg, 99),  # Maximum value of the colorbar
)

License

Released under the Apache 2.0 license:

Copyright (C) 2023-2024 dcmri developers