Magnetic resonance imaging (MRI) has become a standard diagnostic tool in medical assessments. How-
ever, long acquisition times remain one of its main drawbacks. This makes dynamic imaging, as needed
in cineMRI, particularly challenging: motion artefacts and limited data acquisition per time point lead to
an inherently ill-posed reconstruction problem. Typical solution strategies include breath-holding by the
patient to limit induced motion or manual binning of frames according to an external signal.
We propose a method that incorporates motion signal information directly in the image discreti-
sation. Inspired by the deep-image prior [1], invertible residual networks [2], Lipschitz-constrained
networks via Householder reflections, and previous work on motion disentanglement [3], we design a
multi-component neural network-based architecture for image reconstruction. The motion signals, in
this case representative of the breathing and cardiac cycle, are used as conditioning information for the
network. This framework not only allows for retrospective modification and suppression of motion but
also enables the reconstruction of incomplete motion information, such as when no breathing belt is used.
In our presentation, we provide a mathematically sound formulation of the reconstruction problem
within a statistical learning framework, together with general features of the proposed network archi-
tecture. These are complemented by numerical test cases highlighting capabilities such as cine recon-
struction, retrospective motion suppression, feature tracking, displacement extraction, and resolution
independence.
References
[1] Ulyanov, D., Vedaldi, A., Lempitsky, V., Deep image prior, Proceedings of the IEEE conference on computer
vision and pattern recognition, pp. 9446-9454, 2018.
[2] Behrmann, J., Grathwohl, W., Chen, R. T., Duvenaud, D., Jacobsen, J. H., Invertible residual networks,
International conference on machine learning, pp. 573-582, 2019.
[3] Abdullah, A., Holler, M., Kunisch, K., Landman, M. S., Latent-space disentanglement with untrained gen-
erator networks for the isolation of different motion types in video data, International conference on scale
space and variational methods in computer vision, pp. 326-338, 2023.
