Go to unit's home.

Home | DTI | 2007–08 funded proposals | Tom Luo, Pierre-Francois Van de Moortele

Initiatives in Digital Technology: 2007–08 Funded Proposals

Tom Luo, Pierre-Francois Van de Moortele

Optimal B1 Shimming for Ultra High Field Human Scanners

Magnetic Resonance Imaging at 7 Tesla and above (UHF7T+) involves multiple difficult challenges. At higher field, because RF wavelength becomes smaller than the imaged target, destructive and constructive superpositions of interferences, dielectric phenomenon and damping wave patterns can result in uncontrolled variations of signal intensity and tissue contrast through space. Such consequences dramatically narrow the scope of clinical applications of human scanners at UHF7T+. A powerful approach to address this challenge is to use an array of transmit coils whose RF inputs (phase and amplitude) can be individually altered. This technique, named B1 Shim, can significantly improve B1 field homogeneity. However, performing real-time B1 shimming requires fast and reliable signal processing algorithms that can account for large anatomical variability of individual patients and deliver robust optimal solutions on measured data. As an essential component of the UHF7T+ MRI system, fast optimal B1 shimming has attracted significant attention in the last two years from the MRI research community, including the Center for Magnetic Resonance Research at UMN. In contrast, the signal processing community has been slow to respond to this important biomedical application. So far all the B1 shimming techniques reported in MRI literature are ad hoc, sensitive to initialization, and too slow for real time applications. In this pro ject we will use modern convex optimization techniques to develop advanced signal processing tools for real time optimal B1+ shimming, and integrate them into the ultrahigh field human scanners currently under development at CMRR. The outcome of the proposed research is expected to significantly expand the diagnostic capability of fMRI in domains where higher magnetic fields outperform lower fields.