Free-breathing, Highly Accelerated, Single-beat, Multisection Cardiac Cine MRI with Generative Artificial...

Area of Science:

• Cardiovascular Imaging
• Magnetic Resonance Imaging (MRI)
• Medical Technology Assessment

Background:

• Conventional cardiac MRI often requires multiple breath-holds, leading to prolonged scan times and potential patient discomfort.
• Accelerated imaging techniques are crucial for improving patient experience and increasing MRI throughput.
• Deep learning reconstruction methods show promise in enhancing image quality and reducing acquisition time.

Purpose of the Study:

• To develop and evaluate a novel, free-breathing, highly accelerated, multisection, single-beat cine sequence for cardiac MRI.
• To compare the diagnostic accuracy, image quality, and reproducibility of the new sequence against conventional multibreath-hold techniques.

Main Methods:

• A prospective study utilizing a 3-T MRI system with a custom single-beat cine sequence, acquiring data in one heartbeat.
• Images were reconstructed using a deep learning model (REGAIN) for resolution enhancement.
• Comparison involved multibreath-hold k-space-segmented (4.2x acceleration) and free-breathing single-beat (14.8x acceleration) cine sequences, with quantitative and qualitative assessments by expert cardiologists.

Main Results:

• The single-beat sequence reduced scan time dramatically from 2.6 minutes to 0.5 minutes.
• Excellent correlations (r = 0.97-0.99 for LV, r = 0.89-0.98 for RV) were found between single-beat and k-space-segmented parameters.
• High diagnostic (92%) and image quality (98%) ratings were achieved, with excellent scan-rescan reproducibility (ICC = 0.97-1.0).

Conclusions:

• The developed free-breathing, 30-second single-beat cardiac cine MRI provides accurate biventricular measurements.
• This accelerated technique maintains high diagnostic and image quality compared to conventional multibreath-hold methods.
• The single-beat sequence represents a significant advancement in cardiac MRI, improving efficiency and patient experience.