corrField


Logo for corrField

About

Image Version:
20fbff09-2e33-48a7-ba61-83f084ff2ea1
Last updated:
May 3, 2021, 8:08 a.m.
Inputs:
  • Fixed Image  (The target image for registration)
  • Moving Image  (The image that will be warped during registration to the fixed image)
  • Fixed Mask  (Mask of the locations in the fixed image to consider)
  • corrField Configuration  (Configuration object for corrField algorithm where "alpha": regularisation parameter, "beta": intensity weighting, "gamma":scaling factor for soft correspondeces, "delta": step size for mind descriptor, "lambda": regularistion parameter for TPS, "sigma": sigma for foerstner operator, "sigma1": sigma for mind descriptor, "search_radius": maximum search radius for each level, "length": cube-length of non-maximum suppression, "quantisation": quantisation of search step size, "patch_radius": patch radius for similarity seach, "transform": rigid(r)/non-rigid(n).)
Outputs:
  • Results JSON File  (A collection of results of unknown type. Legacy, if possible please use alternative interfaces.)
  • Warped Image  (The moving image warped to the fixed image)
  • Displacement Field  (Displacement vectors to align the moving image to the fixed image)

Model Facts

Summary

Correspondence fields for large motion image registration. This algorithm is an extended PyTorch (GPU support) implementation of the corrField method described in [1] and was developed as part of the work for [2].

Source code:

For further questions or comments please contact us at {hansen,heinrich}@imi.uni-luebeck.de.

[1] Heinrich, Mattias P., Heinz Handels, and Ivor JA Simpson. "Estimating large lung motion in COPD patients by symmetric regularised correspondence fields." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2015.

[2] Hansen, Lasse, and Mattias P. Heinrich. "GraphRegNet: Deep Graph Regularisation Networks on Sparse Keypoints for Dense Registration of 3D Lung CTs." IEEE Transactions on Medical Imaging (2021).

Mechanism

Validation and Performance

Uses and Directions

This algorithm was developed for research purposes only.

Warnings

Common Error Messages

Information on this algorithm has been provided by the Algorithm Editors, following the Model Facts labels guidelines from Sendak, M.P., Gao, M., Brajer, N. et al. Presenting machine learning model information to clinical end users with model facts labels. npj Digit. Med. 3, 41 (2020). 10.1038/s41746-020-0253-3