An optimization framework to maximize signal-to-noise ratio in simultaneous multi-slice body imaging

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Imaging Technology
• Radiological Physics

Background:

• Parallel imaging accelerates MRI scans, reducing motion artifacts in abdominal and pelvic imaging.
• Controlled Aliasing In Parallel Imaging Results IN Higher Acceleration (CAIPIRINHA) offers higher signal-to-noise ratio (SNR) than in-plane parallel imaging.
• Abdominal CAIPIRINHA imaging faces challenges in consistent image quality due to variable coil placement, anatomy, and scan coverage.

Purpose of the Study:

• To develop a mathematical framework for calculating SNR in accelerated MRI acquisitions.
• To optimize sampling patterns for CAIPIRINHA-accelerated abdominal imaging by maximizing local SNR.
• To evaluate the impact of subject-specific factors on image quality and artifacts in simultaneous multi-slice (SMS) accelerated acquisitions.

Main Methods:

• Introduced a mathematical framework to calculate retained SNR for in-plane and simultaneous multi-slice (SMS) accelerated acquisitions.
• Utilized the framework to optimize sampling patterns via non-linear, RF-induced CAIPIRINHA slice shifts, maximizing SNR within a region of interest (ROI).
• Evaluated the framework on 14 healthy subjects, comparing optimized patterns with standard in-plane and linear-shift CAIPIRINHA acceleration.

Main Results:

• Superior-inferior field of view (FOV), coil positioning, and individual anatomy significantly impact SNR (up to 50% variation with coil position, 40% between subjects) and artifacts in SMS acceleration.
• Optimized sampling patterns using non-linear shifts increased SNR by 10-30% for higher acceleration factors.
• Demonstrated the necessity of optimizing sampling patterns for different FOVs and ideally on an individual basis.

Conclusions:

• Consistent image quality and SNR in CAIPIRINHA-accelerated abdominal imaging are challenging due to anatomical and coil placement variability.
• Sampling patterns must be optimized for varying FOVs and ideally personalized for each patient.
• The developed framework enables optimization of sampling patterns for diverse accelerated body MRI acquisitions on an individual basis.