All publications

Conference paper / 2025

Neural Fields for Continuous Periodic Motion Estimation in 4D Cardiovascular Imaging

Simone Garzia, Patryk Rygiel, Sven Dummer, Filippo Cademartiri, Simona Celi, Jelmer M. Wolterink

Statistical Atlases and Computational Models of the Heart (STACOM) · MICCAI 2024 workshop · Published · 2024 conference, 2025 proceedings

A neural network with periodic time encoding models aortic motion throughout the cardiac cycle.
A continuous neural velocity field with periodic time encoding describes aortic motion across the cardiac cycle. Figure 1 · Garzia et al. (2024) · Garzia et al. · CC BY 4.0

Representing motion across the cardiac cycle

This method learns a time-dependent velocity field from a four-dimensional image sequence. Integrating that field produces deformations that can move images, segmentations, or surface meshes through time.

Periodicity is incorporated into the temporal representation and the deformation regularization. Experiments with synthetic sequences, gated CT, and flow MRI investigate how the method can support analysis of moving artery walls.

Cite this work

@inproceedings{neuralmotion2025,
  title = {Neural Fields for Continuous Periodic Motion Estimation in 4D Cardiovascular Imaging},
  author = { Simone Garzia and Patryk Rygiel and Sven Dummer and Filippo Cademartiri and Simona Celi and Jelmer M. Wolterink },
  year = {2025},
  booktitle = {Statistical Atlases and Computational Models of the Heart (STACOM)},
  pages = {378--389},
  doi = {10.1007/978-3-031-87756-8_37},
  eprint = {2407.20728},
  archivePrefix = {arXiv},
  url = { https://doi.org/10.1007/978-3-031-87756-8_37 }
}