Dynamic Regularized Adaptive Cluster Optimization (DRACO) for Quantitative Cardiac Cine MRI in Complex Arrhythmias.

Area of Science:

• Cardiovascular Imaging
• Medical Physics
• Biomedical Engineering

Background:

• Irregular cardiac motion complicates standard cine MRI, potentially making it nondiagnostic.
• Clustering techniques offer a method for cardiac motion binning, with potential for optimization in arrhythmias.

Purpose of the Study:

• To develop an adaptive cluster optimization method, Dynamic Regularized Adaptive Cluster Optimization (DRACO), for irregular cardiac motion.
• To generate time-resolved cine MRI images using the developed DRACO method.

Main Methods:

• Prospective study involving patients with atrial fibrillation, premature ventricular contractions, and sinus rhythm.
• Utilized free-running balanced steady-state free precession (bSSFP) with sorted golden-step acquisition.
• Images reconstructed with DRACO and k-means clustering, compared against a reference real-time sequence.

Main Results:

• DRACO achieved significantly higher image quality scores (≥3) compared to k-means and real-time methods (96% diastolic, 93% systolic/multiphase).
• DRACO demonstrated significantly better image quality, SNR, and CNR than both k-means and real-time methods.
• No significant differences in cardiac function analysis were observed between DRACO and the reference real-time method.

Conclusions:

• DRACO enables high-quality, time-resolved image reconstruction for cardiac MRI.
• The DRACO method shows promise for quantitative cine cardiac MRI in patients with complex arrhythmias like atrial fibrillation.