TR&D 1 Multidimensional MRI The Multidimensional MRI TR&D plans to investigate novel microstructure imaging technologies and biophysi- cal models to better characterize tumor tissue in vivo. The proposed technology will enable investigation into the heterogeneity of tumor microstructure, which has important treatment implications not only during surgery, but also for biopsy guidance and drug delivery. Due to tumor heterogeneity, the tissue microenvironment changes throughout the tumor: it can have locally different diffusivities each associated with its own T1 and T2 relaxation times, along with local changes in cell membrane permeability and blood flow. The lack of non-inva- sive glioma characterization limits the clinician's ability to identify the type of glioma, to define the tumor extent for surgical resection, and to target biopsies. We propose to develop multidimensional MRI sequences and novel image analysis techniques to better characterize tumor tissue microstructure and heterogeneity to pro- vide information not currently available in the clinic. A main focus of our approach is the analysis of multidimensional MRI, which maps the correlation of the MRI signal across multiple dimensions of acquisition parameters by varying these parameters jointly during the ac- quisition. This approach is inspired by multidimensional correlation spectroscopy, which improves differentia- tion of heterogeneous media in the field of NMR. We propose to probe the tumor tissue using multiple com- plementary dimensions and to develop a complete framework for better characterization of tumor type and grade from human in-vivo data. To further explore and visualize such large amounts of data, we will develop a web-based real-time visualization software with parallel computing on a Graphics Processor Unit (GPU). We propose the following specific aims: (1) Multidimensional imaging for improved microstructure quantification, (2) Biophysical models of microscopic water diffusion, perfusion, and permeability. and (3) Interactive web- based multidimensional MR visualization and analysis. Overall, our proposed multidimensional MRI technologies have the potential to provide non-invasive quantifica- tion of multiple tumor features that are not currently available in the clinic. This technology has potential to im- pact several aspects of future surgical workflow, such as diagnosis of brain tumor subtypes and grades, tumor core delineation for surgical and radiotherapy planning, and assessment of treatment response.
Showing the most recent 10 out of 252 publications
