Inverse Problem for Color Doppler Ultrasound-Assisted Intracardiac Blood Flow Imaging

Area of Science:

• Cardiovascular Imaging
• Computational Fluid Dynamics
• Biomedical Engineering

Background:

• Echocardiography is standard for assessing left ventricle (LV) geometry but insufficient for visualizing complex blood flow patterns.
• Characterizing swirling flow in the LV is crucial for understanding cardiac function and disease.
• Current imaging techniques often lack the resolution to fully capture intracardiac hemodynamics.

Purpose of the Study:

• To develop a mathematical framework for reconstructing three-dimensional (3D) left ventricle (LV) blood flow.
• To accurately predict dynamic swirling flow patterns within the LV over a cardiac cycle.
• To overcome limitations of standard echocardiography in visualizing intracardiac blood flow.

Main Methods:

• A mathematical framework based on inverse problems was developed for 3D LV blood flow reconstruction.
• The model integrates incompressible Navier-Stokes equations with velocity data from forward simulations or color Doppler.
• Time-varying LV boundaries were extracted from intensity data to define boundary conditions.

Main Results:

• The proposed reconstruction model significantly reduced local and global errors in reconstructed flow fields.
• Synthetic flow data generated via fluid-structure interaction models were utilized.
• Feasibility was demonstrated in predicting dynamic swirling patterns within the LV.

Conclusions:

• The novel mathematical framework enables accurate 3D reconstruction of LV blood flow.
• This approach enhances the visualization and prediction of dynamic swirling patterns in the LV.
• The method offers potential for improved assessment of cardiac hemodynamics beyond conventional echocardiography.