The objective of this research is to improve the performance of existing nuclear magnetic imaging and spectroscopy methods of applying the concepts of feedback control. These methods are based on observing the fluctuations in nuclear magnetization produced by sequences of RF pulses and, in the case of imaging, applied magnetic field gradients. Conventional imaging and spectroscopy techniques are open-loop processes in that measurements of the magnetization state are used to optimize the pulse parameters between successive RF pulses. This project focuses specifically on contrast enhancement in magnetic resonance imaging. A feedback strategy that automatically adjusts pulse parameters to enhance the contrast between specified chemical species is developed. The control design is verified experimentally on a commercial imager using a sample of water and oil. An adaptive control algorithm tunes the pulse parameters so that the difference between the water and oil magnetization signals is maximized. The significance of this research is twofold. First, it introduces the imaging and spectroscopy communities to the advantages afforded by using feedback. Second, it provides a foundation on which to consider further applications of feedback concepts to nuclear magnetic imaging and spectroscopy.
