Phantom based quality assurance for long-term stability and inter-scanner compatibility in T1, T2 and T2* measurements

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background The quantitative measurement of T1, T2 and T2* relaxation parameters is prone to numerous physical variables related to scanner performance and installation environment. These variables include the presence of eddy currents, main magnetic field inhomogeneity and gradient field performance. Thus, it is important to verify that the measured relaxation parameters are inter-scanner compatible and there are no drift in the measured values over time. Purpose In this study, we produced a phantom for the relaxation parameter measurements. With the phantom, long-time stability and inter-scanner comparison of parameters were studied with preliminary quality assurance protocol. Methods A cylindrical quality assurance phantom with 16 inserts filled with ammonium iron(II) ((NH4)2Fe(SO4)2 · 6H2O) doped gel was produced. The (NH4)2Fe(SO4)2 concentrations were from 0,0119 mmol/g to 0,5498 mmol/g, corresponding to approximate T1 values of 10–1030 ms, T2 values of 16–78 ms and T2* values of 5–61 ms. The phantom was measured quarterly for a one-year period with three scanners from two different vendors to follow the scanner stability and inter-scanner compatibility. The used imaging sequences included T1 MOLLI -based sequence, T2 prep -based sequence, and multi-echo gradient echo sequence for T2* calculation. Simulated ECG signal with 60 beats per minute was used in the scanning to mimic a typical rest heart rate of an adult patient. Results The range of variation between scanners from average relaxation coefficient in T1 was -2% to 4% and in T2 and T2* measurements -7% to 5% and -32% to 18%, respectively. In long-term stability test, the range of variation from average relaxation coefficient in T1 was -1% to 2% and in T2 and T2* measurements -1% to 1% and -15% to 16%, respectively. Conclusion(s) There were significant differences in the measured T1, T2, and T2* relaxation coefficients in both inter-scanner and long-term stability. However, the detected variations were acceptable compared to the typical physiological uncertainties of the measured parameters but have to be taken into account when the reliability of the study is assessed. Abstract Figure. Results and phantom image