Imaging Human Brain Metabolism Exploiting Ultra-High Field MRI

In vivo magnetic resonance spectroscopy has evolved during the last 25 years in terms of localization quality and spatial resolution, acquisition speed, artifact suppression, number of detectable metabolites and quantification precision and has benefited from the significant increase of magnetic field strength that recently became available for in vivo investigations. Today it allows for non-invasive and non-ionizing determination of tissue concentrations and metabolic turn-over rates of more than 20 metabolites and compounds with high spatial resolution in the human brain and has established as an important tool for neurophysiological research. This presentation summarizes our recent work using a human 9.4T whole-body MRI scanner. Advantages and technical challenges of ultra-high field human MRI as well as related hardware (RF coils, B0 shimming), data acquisition (RF pulses, Sequences) and data reconstruction approaches are discussed. The high signal-to-noise ratio (SNR) and the spectral resolution at 9.4T in combination with optimized 1H MRSI acquisition and image reconstruction techniques allows for mapping the spatial distribution of a total of 12 metabolites including neurotransmitters, second messengers and antioxidants in the living human brain. Other measurable substances are involved in energy and membrane metabolism. Visualization of concentration differences between gray and white matter and identification of gyri in metabolic MR brain images becomes possible for the first time. In vivo detection of amino acids in vivo is demonstrated and histograms of the amino acid chemical shift distributions extracted from the protein NMR database are used as a fitting model to quantify them. The SNR and spectral separation at UHF also allows for regional functional metabolic readouts and reveal a modulation of energy metabolites and neurotransmitter concentrations. 31P spectroscopy and spectroscopic imaging of the human brain allow mapping of the spatial distribution of energy metabolites such as ATP and NADH. Finally potential neuroscientific and clinical applications are identified