FusionNet: a frame interpolation network for 4D heart models

Chujie Chang; Shoko Miyauchi; Ken'ichi Morooka; Ryo Kurazume; Oscar Martinez Mozos

cs.CVMar 10, 2026

FusionNet: a frame interpolation network for 4D heart models

Authors: Chujie Chang, Shoko Miyauchi, Ken'ichi Morooka, Ryo Kurazume, Oscar Martinez Mozos

Organizations: Kyushu University, Fukuoka 819-0395, Japan · Okayama University, Okayama 700-8530, Japan · Örebro University, Örebro 701-82, Sweden

Abstract

Cardiac magnetic resonance (CMR) imaging is widely used to visualise cardiac motion and diagnose heart disease. However, standard CMR imaging requires patients to lie still in a confined space inside a loud machine for 40-60 min, which increases patient discomfort. In addition, shorter scan times decrease either or both the temporal and spatial resolutions of cardiac motion, and thus, the diagnostic accuracy of the procedure. Of these, we focus on reduced temporal resolution and propose a neural network called FusionNet to obtain four-dimensional (4D) cardiac motion with high temporal resolution from CMR images captured in a short period of time. The model estimates intermediate 3D heart shapes based on adjacent shapes. The results of an experimental evaluation of the proposed FusionNet model showed that it achieved a performance of over 0.897 in terms of the Dice coefficient, confirming that it can recover shapes more precisely than existing methods. This code is available at: https://github.com/smiyauchi199/FusionNet.git

Figures & tables

	Frame 2	Frame 4	Frame 6	Frame 8	Frame 10	Average
FusionNet	.897 $\pm$ .005	.879 $\pm$ .006	.877 $\pm$ .007	.928 $\pm$ .003	.905 $\pm$ .006	.897 $\pm$ .019
ConvLSTM [ 9 ]	.884 $\pm$ .003	.867 $\pm$ .005	.854 $\pm$ .006	.912 $\pm$ .003	.892 $\pm$ .007	.881 $\pm$ .020
U-Net [ 12 ]	.892 $\pm$ .004	.875 $\pm$ .004	.871 $\pm$ .007	.922 $\pm$ .003	.899 $\pm$ .008	.892 $\pm$ .018
Bilinear [ 4 ]	.821 $\pm$ .008	.831 $\pm$ .008	.813 $\pm$ .008	.915 $\pm$ .004	.890 $\pm$ .005	.854 $\pm$ .041

Table 1: Comparison with previous methods using the average Dice coefficient from 7-fold cross validation.

Figure 2: Cross-section of the generated voxel model corresponding to the estimated frame (e.g. frames 2 and 8) and the difference image between the model and the ground truth. The last column, WP, indicates the number of white pixels in the difference image. The results of our FusionNet contain the lower values.

Panel of Figure 3 — Figure 3: (a) Average volume change in ground truth 4D cardiac models. (b) Relationship between the average Dice coefficients and the input frame interval for FusionNet and the three existing methods (* p < .05).

4D Reconstruction

4D reconstruction recovers changing 3D scenes, shapes, and motion over time from images, videos, or other sparse measurements.

See all 139 papers

CineMesh4D: Personalized 4D Whole Heart Reconstruction from Sparse Cine MRI
May 13, 2026Xiaoyue Liu, Xiaohan Yuan, Mark Y Chan +23D Mesh Generation 4D Reconstruction
Department of Biomedical Engineering, National University of Singapore, Singapore · School of Automation, Southeast University, Nanjing, China · Department of Medicine, National University of Singapore, Singapore +1
End-to-End 4D Heart Mesh Recovery Across Full-Stack and Sparse Cardiac MRI
Sep 15, 2025Yihong Chen, Jiancheng Yang, Deniz Sayin Mercadier +34D Reconstruction 3D Reconstruction
School of Computer and Communication Science, EPFL · ELLIS Institute Finland, Aalto University, Finland · University of North Carolina at Charlotte +1
NISF++: Geometrically-grounded implicit representations of 3D+time cardiac function from 2D short- and long-axis MR views
Aug 1, 2026Nil Stolt-Ansó, Maik Dannecker, Steven Jia +24D Reconstruction Implicit Neural Representations
Munich Center for Machine Learning, Technical University Munich, Germany · School of Computation, Information and Technology, Technical University Munich, Germany · TUM University Hospital, Technical University Munich, Germany +2

May 13, 2026J/K move · Enter open · S save

CineMesh4D: Personalized 4D Whole Heart Reconstruction from Sparse Cine MRI

Xiaoyue Liu, Xiaohan Yuan, Mark Y Chan, Ching-Hui Sia+1

Department of Biomedical Engineering, National University of Singapore, Singapore · School of Automation, Southeast University, Nanjing, China · Department of Medicine, National University of Singapore, Singapore+1

3D Mesh Generation 4D Reconstruction 3D Reconstruction Image Reconstruction Sparse-View 3D Reconstruction Magnetic Resonance Imaging+1

Open paper AskPDF arXiv

Accurate 3D+t whole-heart mesh reconstruction from cine MRI is a clinically crucial yet technically challenging task. The difficulty of this task arises from two coupled factors: inherently sparse sampling of 3D cardiac anatomy by 2D image slices and the tight coupling between cardiac shape and motion. Current cardiac image-to-mesh approaches typically reconstruct only a subset of cardiac chambers or a single phase of the cardiac cycle. In this work, we propose CineMesh4D, a novel end-to-end 4D (3D+t) pipeline that directly reconstructs patient-specific whole-heart mesh from multi-view 2D cine MRI via cross-domain mapping. Specifically, we introduce a differentiable rendering loss that enables supervision of 3D+t whole-heart mesh from multi-view sparse contours of cine MRI. Furthermore, we develop a dual-context temporal block that fuses global and local cardiac temporal information to capture high-dimensional sequential patterns. In quantitative and qualitative evaluations, CineMesh4D outperforms existing approaches in terms of reconstruction quality and motion consistency, providing a practical pathway for personalized real-time cardiac assessment. The code will be publicly released once the manuscript is accepted.

Abstract

4D Reconstruction

Explore similar work

CineMesh4D: Personalized 4D Whole Heart Reconstruction from Sparse Cine MRI

End-to-End 4D Heart Mesh Recovery Across Full-Stack and Sparse Cardiac MRI

NISF++: Geometrically-grounded implicit representations of 3D+time cardiac function from 2D short- and long-axis MR views