PROJECT SUMMARY MR spectroscopic imaging (MRSI) is a powerful, non-invasive method to map the metabolic profile of human brain. The intrinsic chemical specificity of MRSI can reveal altered metabolism in regions that would appear non-pathologic on standard MRI. MRSI has successfully been used for brain tumor classification and infiltration, planning of radiation therapy and surgery and has helped to characterize a range of neurodegenerative diseases such as multiple sclerosis, Alzheimer's disease and epilepsy. Despite its merits and successes, MRSI is not a widespread imaging modality because the method is hampered by the long- standing problem of magnetic field inhomogeneity, which leads to lower-quality and thus less-reproducible data. Incomplete water and lipid suppression can compromise data quality even further. Over the last decade we have developed multi-coil (MC) shimming as an alternative to conventional spherical harmonics based shimming. By controlling the current in a matrix of generic, circular DC coils the magnetic field homogeneity can be improved to less than 0.1 ppm across the entire human brain. In this renewal application we propose to apply our extensive experience with MC field manipulation and the superior performance of MC shimming to achieve consistent and reliable MRSI performance across the human brain. Aim 1 is focused on the implementation of MC shimming and its integration with ultra-fast, EPI-based MRSI. The MC technology will be modified and extended to generate local crusher gradients for effective lipid suppression. Aim 2 is concerned with the validation of MC-augmented MRSI on the human brain. Besides comparisons between MC and conventional shimmed MRSI, Aim 2 will also establish intra- and inter-subject, as well as test-retest reproducibility. In addition, the MC setup optimized at 4 T will be transferred to 3 T and 7 T, without modifications, to demonstrate compatibility of MC-based shimming with clinical MR systems. The implementation and validation of MC-augmented MRSI is considered complete when >90% of MRSI voxels on all slices, all subjects and all scanners adhere to stringent data quality requirements. While all neurological disorders and pathologies can benefit from high-fidelity MRSI, Aim 3 focuses on obtaining high-quality, reliable MRSI on tumor patients. The presence of metallic implants in many post-surgery tumor patients, in addition to tumor-related pathology, compromises the magnetic field homogeneity. MC shimming will be used to demonstrate that high-fidelity MRSI can be achieved on a challenging, clinical subject population.