Automatic estimation of aortic and mitral valve displacements in dynamic CTA with 4D graph-cuts

Area of Science:

• Cardiovascular Imaging
• Medical Image Analysis
• Computational Anatomy

Background:

• Accurate segmentation of mitral and aortic valves in dynamic cardiac imaging is crucial for assessing cardiac function and planning interventions.
• Current automatic segmentation methods face challenges due to rapid valve motion and variable visibility throughout the cardiac cycle.

Purpose of the Study:

• To develop and evaluate a processing pipeline for tracking the displacement of the aortic and mitral valve annuli from high-resolution 4D-CTA.
• To assess the accuracy of the proposed method compared to manual annotations.
• To demonstrate the impact of accurate valvular plane detection on derived functional parameters.

Main Methods:

• A processing pipeline utilizing statistical shape modeling and graph-cuts energy minimization for dynamic separation of cardiac structures (left ventricle, left atrium, aorta).
• Tracking of aortic and mitral valve annulus displacement from electrocardiography-gated 4D-CT angiography (4D-CTA) data.
• Evaluation on 15 4D-CTA datasets with comparison against manual anatomical landmark annotations.

Main Results:

• The proposed method achieved a mean agreement distance of 2.52±1.06 mm for the mitral annulus and 2.00±0.69 mm for the aortic valve annulus.
• Demonstrated the influence of precise valvular plane detection on the accuracy of derived functional parameters, including ejection fraction, global longitudinal strain, and valve excursions.

Conclusions:

• The developed processing pipeline offers an accurate approach for tracking mitral and aortic valve annulus displacement in 4D-CTA.
• This method has the potential to improve the reliability of functional parameter extraction in cardiac imaging analysis.
• Accurate valve annulus tracking is essential for robust quantification of cardiac mechanics and clinical decision-making.