Abstract
Cardiac motion is a fundamental indicator of myocardial function, yet its regional assessment in clinical practice still relies primarily on subjective visual interpretation of wall motion. Quantities derived from temporal derivatives of the motion field, such as velocity and acceleration, could provide objective descriptors of regional myocardial function while also supporting patient-specific biomechanical models and cardiac digital twins. However, cine Magnetic Resonance Imaging (MRI) captures cardiac structures only at a finite number of, typically 10-25, time points over a complete cardiac cycle, limiting direct access to con- tinuous kinematic information. We propose a framework that transforms standard cine MRI into an analytical representation of cardiac motion. Cardiac geometry is first encoded into a compact latent space, where temporal dynamics are modelled by a variational autoencoder and represented analytically using a polynomial formulation. This yields a continuous expression of cardiac motion throughout the cardiac cycle, enabling direct analytical differentiation of velocity fields and other derivative-based quantities while avoiding pointwise trajectory fitting. Experiments on 525 healthy subjects from the UK Biobank demonstrate the feasibility of reconstructing a continuous analytical description of cardiac motion from standard cine MRI, providing a foundation for quantitative analysis of cardiac motion.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/STACOM2026_046.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/profile?id=~Yasmine_Boukhriss1
BibTex
@InProceedings{BouYas_Continuous_MICCAISAT2026,
author = { Boukhriss, Yasmine AND Seale, Thalia AND Banerjee, Abhirup},
title = { { Continuous Volumetric Cardiac Motion Modelling for Analytical Derivatives from Cine MRI } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17267},
month = {pending},
page = {pending}
}