Pairwise registration for grouped cardiac MR images

Note: Please read the DeepReg Demo Disclaimer.

Source Code

This demo uses the grouped dataset loader to register intra-subject multi-sequence cardiac magnetic resonance (CMR) images.

Author

DeepReg Development Team

Application

Computer-assisted management for patients suffering from myocardial infraction (MI) often requires quantifying the difference and comprising the multiple sequences, such as the late gadolinium enhancement (LGE) CMR sequence MI, the T2-weighted CMR. They collectively provide radiological information otherwise unavailable during clinical practice.

Instruction

  • Install DeepReg;

  • Change current directory to the root directory of DeepReg project;

  • Run demo_data.py script to download all the CMR dataset in a zip file. The script also splits the data into train, val and test sets re-samples all the images to an isotropic voxel size.

python demos/grouped_mr_heart/demo_data.py

  • Call deepreg_train from command line. The following example uses a single GPU and launches the first of the ten runs of a 9-fold cross-validation, as specified in the ``dataset` section <./grouped_mr_heart_dataset0.yaml>`_ and the ``train` section <./grouped_mr_heart_train.yaml>`_, which can be specified in seperate yaml files;

deepreg_train --gpu "0" --config_path demos/grouped_mr_heart/grouped_mr_heart.yaml --log_dir grouped_mr_heart

  • Call deepreg_predict from command line to use the saved ckpt file for testing on the data partitions specified in the config file, a copy of which will be saved in the [log_dir]. The following example uses a pre-trained model, on CPU. If not specified, the results will be saved at the created timestamp-named directories under /logs.

deepreg_predict --gpu "" --config_path demos/grouped_mr_heart/grouped_mr_heart.yaml --ckpt_path demos/grouped_mr_heart/dataset/pre-trained/weights-epoch500.ckpt --save_png --mode test

Pre-trained Model

A pre-trained model will be downloaded after running demo_data.py and unzipped at dataset folder under the demo folder. This pre-trained model will be used by default with deepreg_predict. Run the user-trained model by specifying with --ckpt_path the location where the ckpt files will be saved, in this case (specified by deepreg_train as above), /logs/grouped_mr_heart/.

Data

This demo uses CMR images from 45 patients, acquired from the MyoPS2020 challenge held in conjunction with MICCAI 2020.

Tested DeepReg version

Last commit: 74e7b1f749d0df1c140494eba0204f0edd1d7b1e

Contact

Please raise an issue.

Reference

[1] Xiahai Zhuang: Multivariate mixture model for myocardial segmentation combining multi-source images. IEEE Transactions on Pattern Analysis and Machine Intelligence (T PAMI), vol. 41, no. 12, 2933-2946, Dec 2019. link.

[2] Xiahai Zhuang: Multivariate mixture model for cardiac segmentation from multi-sequence MRI. International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.581-588, 2016.