Feedback Control of the Nuclear Magnetization State: Experimental Results

Conventional MRI is an open-loop process in the sense that measurements of the magnetization state are not utilized until after the completion of a pulse sequence. Optimization of pulse parameters occurs before the pulse sequence is transmitted. In a previous paper, we investigated feedback control of the nuclear magnetization state [1]. This strategy facilitates the optimization of the pulse parameters during the imaging sequence. A systematic method for modeling and controlling the nuclear magnetization state was presented. The method was illustrated by a theoretical example in which the angle between the bulk magnetization and the axis of an applied static magnetic field is controlled using feedback. In this paper, results from a series of experiments used to test the theoretical predictions are presented. Two groups of experiments are performed using a commercial imager and a small sample of distilled water. In the first group of experiments the control objective is to regulate the magnetization orientation to a constant reference angle. These experiments are used to compare the performance of open and closed-loop systems. The experimental results show that a closed-loop system produces the desired orientation. In contrast, an open-loop input results in an appreciable error between the desired and measured orientation. A second group of experiments shows that a closed-loop system can force the magnetization orientation to track a desired trajectory. In both groups, the closed-loop experimental and simulation results are in excellent agreement. The application of feedback control to MRI and spectroscopy is discussed.