The presentation will introduce computer simulations of classical MR physics, incorporating state-of-the-art MRI scanner hardware and some of their often-inevitable imperfections.
Realistic computer simulations of MRI experiments rely on a multitude of different physical models. The validity of the models determines agreement of simulation and real experiments. In this talk MRI physics simulations are discussed, which are based on classical spin physics given by the Bloch Equations. In order to accurately simulate the time evolution of the MR signal, the Bloch equations are numerically integrated for a large ensemble of isochromats. Subsequent averaging of the isochromats provides an estimate of the MR signal [1,2].
Modeling scanner imperfections affects the driving fields of the Bloch Equation, i.e. the RF pulses and the gradient waveforms of the MRI sequence. The main effects of gradient hardware imperfections can be well described by spatially non-linear encoding fields as well as time-dependent system responses, the so-called eddy currents. It will be shown that both effects can be easily integrated in numerical solutions of the Bloch equation, if a flexible software design of the imaging sequence is given. Such an environment is provided by the JEMRIS simulation environment [2], which will serve for most of the examples shown in this lecture. The accuracy of such simulations is determined by the validity of the spatio-temporal model of the gradient imperfection, which for instance can be modelled with according low frequency EM solvers [3]. However, this lecture will not present such EM simulations, but show how to easily incorporate the results in Bloch-Equation-based simulations of the MR signal.
Several modern MR physics simulators enable realistic simulation of multiple RF receivers and transmitters, e.g. to simulate parallel imaging acceleration or parallel transmission [1,2]. The approach presented here will take the spatial sensitivity distribution of a specific RF transmit or receive coil array as input for the MR simulator. Again, the input may be the result of a foregoing EM simulation or can be taken from a real MR experiment. Since MRI simulations with a large number of transceivers is much less expensive than an according experiment with real hardware, such investigations can provide meaningful pilot studies supporting the hardware design. The lecture will show an example of validating large-tip angle parallel transmit pulse design by means of MR Physics simulations.
With respect to simulations of gradient hardware imperfections and/or multiple receivers, it is important to provide appropriate interfaces for the MR simulator to state-of-the-art image reconstruction packages such as BART or Gadgetron [4,5]. In this way, the exact ground truth (the numerical phantom) can be compared to the reconstructed image and causes of deviations can be systematically studied, e.g. by specifically including or excluding modeled system imperfections.
Finally, it is of interest to compare MR simulations with system imperfections to real MR experiments. Again, it is helpful to have an appropriate interface which enable to execute exactly the same sequence on the scanner, which was used for the simulation. A convenient and powerful realization is given by combining the JEMRIS simulator with the Pulseq framework [6], which will be shown by example.
[1] Cao, Z., Oh, S., Sica, C. T., McGarrity, J. M., Horan, T., Luo, W., & Collins, C. M. (2014) Bloch-based MRI system simulator considering realistic electromagnetic fields for calculation of signal, noise, and specific absorption rate. Magnetic Resonance in Medicine, 72(1), 237–247.
[2] Stöcker, T., Vahedipour, K., Pflugfelder, D., & Shah, N. J. (2010). High-performance computing MRI simulations. Magnetic Resonance in Medicine, 64(1), 186–93.
[3] F. Liu, S. Crozier, A distributed equivalent magnetic current based FDTD method for the calculation of E-fields induced by gradient coils, J. Magn. Reson. 169 (2004) 323–327.
[4] Martin Uecker, Frank Ong, Jonathan I Tamir, Dara Bahri, Patrick Virtue, Joseph Y Cheng, Tao Zhang, and Michael Lustig. Berkeley Advanced Reconstruction Toolbox. Annual Meeting ISMRM, Toronto 2015, In Proc. Intl. Soc. Mag. Reson. Med. 23:2486.
[5] Hansen, M., & Sørensen, T. (2013). Gadgetron: an open source framework for medical image reconstruction. Magnetic Resonance in Medicine, 69(6):1768-1776.
[6] Layton, K. J., Kroboth, S., Jia, F., Littin, S., Yu, H., Leupold, J., Nielsen, J-F., Stöcker, T., Zaitsev, M. (2017). Pulseq: A rapid and hardware-independent pulse sequence prototyping framework. Magnetic Resonance in Medicine, 77(4), 1544–1552.