Related Experiment Video
Updated: Oct 5, 2025
Human Fetal Blood Flow Quantification with Magnetic Resonance Imaging and Motion Compensation
Published on: January 7, 2021
Probabilistic Modeling for Image Registration Using Radial Basis Functions: Application to Cardiac Motion Estimation
Insights
This study introduces a novel probabilistic framework for estimating cardiac motion using compact support radial basis functions and deep learning. The method enhances accuracy and smoothness in cardiac image registration for better cardiovascular disease assessment.
Area of Science:
- Medical imaging analysis
- Computational cardiovascular dynamics
- Machine learning in healthcare
Background:
- Cardiovascular diseases (CVDs) are a leading cause of mortality, necessitating accurate assessment of cardiac dynamics.
- Cardiac motion estimation is crucial for clinical tasks and understanding cardiac cycle variations.
- Existing image registration methods face challenges in accurately capturing complex cardiac deformations.
Purpose of the Study:
- To propose a probabilistic framework for cardiac motion estimation using compact support radial basis functions (CSRBFs).
- To develop deep learning models for learning probabilistic coefficients of CSRBFs for image deformation.
- To evaluate the framework's performance in cardiac motion estimation and myocardial strain calculation.
Main Methods:
- A generative model using variational inference and convolutional neural networks (CNNs) to learn CSRBF coefficients.
- Two CNN architectures: one for fixed control points and another for drifting control points.
- Derivation of bending energy (BE) for regularization of coefficients within a variational bound.
Main Results:
- The proposed framework demonstrated superior performance compared to state-of-the-art methods in cardiac motion estimation.
- Achieved enhanced deformation smoothness and registration accuracy in experiments.
- Validated on 1409 cardiac MR image slice pairs from public datasets.
Conclusions:
- The probabilistic framework effectively estimates cardiac motion and myocardial strain.
- The deep learning-based approach offers improved accuracy and smoothness in cardiac image registration.
- This method holds potential for advancing cardiovascular disease diagnosis and monitoring.
Abstract:
Cardiovascular diseases (CVDs) are the leading cause of death, affecting the cardiac dynamics over the cardiac cycle. Estimation of cardiac motion plays an essential role in many medical clinical tasks. This article proposes a probabilistic framework for image registration using compact support radial basis functions (CSRBFs) to estimate cardiac motion. A variational inference-based generative model with convolutional neural networks (CNNs) is proposed to learn the probabilistic coefficients of CSRBFs used in image deformation. We designed two networks to estimate the deformation coefficients of CSRBFs: the first one solves the spatial transformation using given control points, and the second one models the transformation using drifting control points. The given-point-based network estimates the probabilistic coefficients of control points. In contrast, the drifting-point-based model predicts the probabilistic coefficients and spatial distribution of control points simultaneously. To regularize these coefficients, we derive the bending energy (BE) in the variational bound by defining the covariance of coefficients. The proposed framework has been evaluated on the cardiac motion estimation and the calculation of the myocardial strain. In the experiments, 1409 slice pairs of end-diastolic (ED) and end-systolic (ES) phase in 4-D cardiac magnetic resonance (MR) images selected from three public datasets are employed to evaluate our networks. The experimental results show that our framework outperforms the state-of-the-art registration methods concerning the deformation smoothness and registration accuracy.
