Next Generation Chemical Exchange saturation transfer MRI

磁共振成像（MRI）是一种无创成像技术，它结合了最佳的软组织对比度和高空间分辨率。因此，它是许多解剖学问题的首选方法。除了软组织形态学成像的优异性能外，MRI对功能过程也具有很高的敏感性，并将用于例如灌注和神经元激活的成像。然而，许多疾病与细胞功能直接相关，并以代谢过程为特征。与体内光谱磁共振也能够研究某些代谢过程，但这种技术遭受相当低的灵敏度，导致低空间分辨率和部分体积问题。CEST MRI利用从代谢物分子到高浓度组织水的化学交换饱和转移，能够将代谢物的敏感性提高上千倍甚至更多。然而，CEST MRI的广泛实际应用仍然存在一些挑战。这些问题包括用于CEST定量的图像序列的长扫描时间，射频饱和的基本技术限制，信息编码的最佳分子饱和度以及分子信息定量提取的复杂后处理。该项目旨在结合磁共振研究的新先进方法和最近的数学方法，实现代谢条件的快速定量成像的重要一步。目前主要有两大研究领域，一方面是从根本上缩短扫描时间，另一方面是提高特定分子饱和度以增强全身系统的灵敏度。为了减少扫描时间，将平行成像与子采样和压缩感知方法相结合进行重建。这种组合对于具有内在相关性的图像序列似乎特别成功，因为它被期望用于量化的CEST光谱。它的目标是将数据采集速度提高十倍，甚至更多，图像质量与传统扫描相似。在进一步的步骤，定量分析与重建的整合进行了探讨。近年来，最优控制被用于设计符合全身系统技术限制的射频切片选择性激励和再聚焦脉冲，其性能比目前最先进的脉冲要好得多。考虑到交换速率、共振偏移和溢出效应，该方法现在应该首次用于优化不同代谢物质子的光谱选择性饱和。如果所有目标都能成功实现，该建议将为MRI独特的形态、功能和代谢成像模式的发展做出决定性的贡献，并为MRI在未来更广泛的代谢生物标志物成像和精准医学的临床应用铺平道路。

Magnetic resonance imaging (MRI) is a noninvasive imaging technique that combines the best soft tissue contrast with high spatial resolution. It is therefore the preferred method for many anatomical questions. Additional to the excellent properties in morphological imaging of soft tissues MRI has also a high sensitivity for functional processes and will be used for example for imaging of perfusion and neuronal activation. However, many diseases are directly connected to cellular function and characterized by metabolic processes. With in-vivo spectroscopy magnetic resonance is also able to investigate certain metabolic processes but this technique suffer from a rather low sensitivity leading to low spatial resolution and partial volume problems. CEST MRI uses the chemical exchange saturation transfer from metabolite molecules to tissue water with its high concentration and is able to enhance the metabolite sensitivity by factors of up to thousand or even more. However, for a broad practical application of CEST MRI some challenges still exist. These are long scanning times for image series used in CEST quantification, basic technical limitations for RF saturation, optimal molecular saturation for information encoding and a complex post-processing for quantitative extraction of molecular information. This projects aims to combine new advanced methods in MR research and recent mathematical approaches to achieve a major step forward for fast and quantitative imaging of metabolic conditions. There are two major research areas, on the one hand the fundamental reduction of scanning time and on the other hand the improvement of the specific molecular saturation for sensitivity enhancement at whole body systems. For the reduction of the scanning time a combination of parallel imaging with subsampling and com-pressed sensing methods for reconstruction will be explored. Such a combination seems to be particular successful for image series with inherent correlation as it is expected for CEST spectra used for quantification. It is aimed to accelerate data acquisition by a factor of ten or even more with similar image quality to conventional scanning. In a further step the integration of quantitative analysis into reconstruction is explored. Optimal control was recently used to design RF slice selective excitation and refocusing pulses complying technical limitations of whole body systems with much better properties than state of the art pulses. This method should now be used for the first time to optimize spectral selective saturation of different metabolite protons considering exchange rate, resonance offset and spill-over effects. If all aims are successfully achieved the proposal can make a decisive contribution to the development of MRI to unique morphological, functional and metabolic imaging modality and prepare the way of MRI for a broader future clinical application in metabolic biomarker imaging and precision medicine.