Accelerated cardiac perfusion imaging using k-t SENSE with SENSE training

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Imaging
• Biomedical Engineering

Background:

• Accelerated MRI acquisition using parallel imaging and k-space undersampling is crucial for dynamic imaging.
• k-t SENSE resolves aliasing by leveraging spatiotemporal correlations in dynamic images.
• Improving temporal fidelity in contrast-enhanced myocardial perfusion imaging at high accelerations remains a challenge.

Purpose of the Study:

• To present a modified k-t SENSE reconstruction approach for enhanced temporal fidelity.
• To improve the quality of first-pass, contrast-enhanced myocardial perfusion images acquired at high acceleration rates.
• To investigate the efficacy and limitations of the proposed method using simulations and in vivo experiments.

Main Methods:

• A modified k-t SENSE reconstruction technique was developed.
• Parallel imaging was applied to training data to enhance spatial resolution.
• The method was evaluated using computer simulations and in vivo experiments.

Main Results:

• Accurate representations of dynamic signal intensities were achieved at a net acceleration of 5.8 (k-t factor = 8, training profiles = 11).
• The modified k-t SENSE approach demonstrated improved temporal fidelity for myocardial perfusion imaging.
• Noise amplification was identified as a potential limitation.

Conclusions:

• The proposed modified k-t SENSE method effectively improves temporal fidelity in accelerated myocardial perfusion MRI.
• Enhancing spatial resolution of training data is a viable strategy for high-acceleration dynamic imaging.
• Further investigation into noise reduction strategies is warranted for clinical translation.