3D+time left ventricular strain by unwrapping harmonic phase with graph cuts.

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Cardiovascular Research

Background:

• Three-dimensional left ventricular strain analysis is crucial for understanding cardiac function.
• Current methods using tagged MRI (tMRI) and manual phase unwrapping (mSUP) are time-consuming and not clinically applicable.
• Accurate measurement of myocardial displacement is essential for precise strain calculation.

Purpose of the Study:

• To develop and validate an automated graph cuts based phase unwrapping method (caSUP) for computing 3D+time cardiac strain.
• To compare the efficiency and accuracy of caSUP against manual unwrapping (mSUP) and a feature-based (FB) method.
• To establish a clinically feasible approach for myocardial strain analysis using tMRI.

Main Methods:

• Developed an automated graph cuts based phase unwrapping algorithm (caSUP) for myocardium displacement measurement.
• Utilized a prolate spheroidal B-spline (PSB) method for 3D strain reconstruction from tMRI data.
• Optimized caSUP parameters using 8 human studies and validated with 32 human studies, comparing results with mSUP and FB methods.

Main Results:

• The automated caSUP method significantly reduced processing time for cardiac strain analysis (6 minutes vs. 2 hours for mSUP).
• Strains computed using caSUP were comparable to those obtained from the manual method (mSUP).
• The caSUP method demonstrated feasibility for accurate 3D+time cardiac strain computation.

Conclusions:

• The automated graph cuts based phase unwrapping (caSUP) method provides an efficient and accurate approach for 3D cardiac strain analysis from tMRI.
• caSUP offers a significant improvement in speed, making it suitable for clinical applications.
• This automated method enhances the clinical utility of tMRI for cardiovascular assessment.