### Description of the supplementary material ###


## 1) Videos

See the folder "videos" for gif images animating the results of Section 3 of the paper and of the appendix.

phantom_bothcounts.gif # Phantom data using both motion triggers
phantom.gif # Phantom data using one motion trigger
case01.gif # Four-chamber-view experiment
case09.gif # Short-axis-view experiment
case02.gif # Second Four-chamber-view experiment
case07.gif # Second Short-axis-view experiment
case01_countnoise.gif # Four-chamber-view experiment, based on perturbed motion trigger
case09_countnoise.gif # Short-axis-view experiment, based on perturbed motion trigger

In addition to this, we also provide videos showing the best result that was obtained when repeating the method with 20 different seeds. See the corresponding files "*_opt.gif" for these result.

In each of the animations, the following is depicted:

Ground truth containing both motion types          | Generated images containing both motion types       | Difference image
Ground truth containing only cardiac motion        | Generated images containing only cardiac motion     | Difference image
Ground truth containing only respiratory motion    | Generated images containing only respiratory motion | Difference image


## 2) Code

To reproduce the results of Section 3 of the paper, run the following:

# Phantom data using both motion triggers:

python testscript_phantom_bothcounts.py # Produces the result 20 times for 20 different seeds

# Phantom data using one motion trigger:

python testscript_phantom.py # Produces the result 20 times for 20 different seeds

# Four-chamber-view experiment:

python testscript_real_case01.py # Produces the result 20 times for 20 different seeds

# Short-axis-view experiment:

python testscript_real_case09.py # Produces the result 20 times for 20 different seeds

# Four-chamber-view experiment based on perturbed motion trigger:

python testscript_real_case01_countnoise.py # Produces the result 20 times for 20 different seeds

# Short-axis-view experiment based on perturbed motion trigger:

python testscript_real_case09_countnoise.py # Produces the result 20 times for 20 different seeds

# Evaluate all results and produce the figures in the paper

python create_paper_results.py



