Hadamard NMR imaging with slice selection.

Stochastic NMR imaging is one of the less common NMR imaging techniques. Nevertheless, stochastic rf excitation is characterized by some remarkable features: the rf excitation power is at least two orders of magnitude lower in comparison to conventionally pulsed NMR imaging schemes. Thus, the technique is of interest for imaging of large objects. The systematic noise inherent in images obtained with random noise excitation has been eliminated by using pseudorandom noise together with Hadamard transformation for data evaluation. Data acquisition times are comparable to those of ultrafast imaging techniques. For slice selection, z magnetization is destroyed outside the slice region with specially designed low power pulses. Thus, gradient switching times are only limited by T1 and not T2*. Images are reconstructed by the backprojection algorithm. We have set up a stochastic imaging procedure on a conventional Bruker MSL 300 spectrometer, and have drawn a comparison between images obtained by the pseudorandom noise excitation and by conventional Fourier imaging.