Comparison of SNR Assessment Techniques for Routine Multi-Element RF Coil QC Testing

Abstract

In the UK, the Institute of Physics and Engineering in Medicine (IPEM) provides guidance to Medical Physics departments on appropriate Quality Control (QC) tests to evaluate MRI scanners used in routine clinical practice. The method recommended for the rigorous annual assessment of the SNR produced by RF coils uses a sequence and regions of interest (ROIs) recommended by IPEM, and calculates SNR through a subtraction calculation (IPEM recommended Method). This method was compared to alternative methods proposed by NessAiver at the 2019 American Association of Physicists in Medicine meeting in their talk on RF coil testing. Comparisons were completed for sequences and regions of interest (ROIs) recommended by IPEM and NessAiver. Testing was performed at 1.5 T using the scanner’s integrated body coil and at 3.0 T using a peripheral Head/Neck coil. Calculation of SNR using the mean of the background noise, assessed using the NessAiver recommended sequence and ROIs (NessAiver Noise-Average Method), typically offered the lowest variability in SNR results. Additionally, the SNR results produced by the IPEM Recommended Method were less repeatable than those from the NessAiver Noise-Average Method (p<0.001). Furthermore, the significance level to which a simulated reduction in SNR could be detected using the IPEM Recommended Method (p<0.001) was less than with the NessAiver Noise-Average Method (p≥0.0031). Finally, when comparing the test duration of each method, use of the NessAiver Noise-Average Method results in a 90% reduction in acquisition time per SNR result, when compared to the IPEM Recommended Method.