Strain analysis from 4-D cardiac CT image data.

Area of Science:

• Cardiology
• Medical Imaging
• Biomechanical Engineering

Background:

• Myocardial strain analysis is crucial for evaluating regional myocardial dysfunction.
• Existing strain analysis techniques are predominantly used with echocardiography and MR imaging.
• Cardiac CT data presents unique challenges for strain analysis, including sparse deformation clues and low temporal resolution.

Purpose of the Study:

• To develop and validate an algorithm for assessing left ventricular (LV) mechanical function using cardiac CT data.
• To overcome limitations of applying strain analysis to cardiac CT.
• To enable clinical application of myocardial strain analysis with CT.

Main Methods:

• Developed a deformable LV model incorporating myocardium and blood pool, accounting for elasticity and incompressibility.
• Utilized image intensities of trabeculae and papillary muscles, along with border edges, via optical flow to extract 3-D velocities.
• Validated the algorithm against 2-D speckle tracking analysis and expert visual scores.

Main Results:

• The algorithm successfully derived 3-D velocities, strains, and rotational values from cardiac CT data.
• Results from normal patients showed high correlation with established literature values.
• Validation confirmed the algorithm's accuracy compared to existing methods and expert assessment.

Conclusions:

• Cardiac CT data can be effectively utilized for myocardial strain analysis.
• The proposed algorithm demonstrates feasibility for clinical application in assessing LV mechanical function.
• This technique offers a new avenue for non-invasive cardiac assessment using CT imaging.