NMR spectroscopy has two main application fields: High Resolution Spectroscopy (HRS), for biochemical characterization, and Magnetic Resonance Imaging (MRI), for medical diagnosis.
HRS is strategic for the modern chemical and pharmaceutical industry. Multidimensional (MD) NMR, one of the leading techniques, has strong computational demands, forcing users to off-line processing. This hinders complete exploitation of costly experimental data, ever increasing with new-generation spectrometers. Furthermore, conventional systems do not allow multidimensional Maximum Enthropy Reconstruction (MER) in an economical time scale. MER would cut MD acquisition times by a factor of two or more, thus enabling the study of compounds so far considered insufficiently stable to be amenable to NMR characterization.
MRI is an important non invasive diagnostic tool. It involves processing huge matrices, for both image reconstruction and image analysis, and new techniques are going to push computational resources even further. Also, MRI users need intensive processing for analysis and interpretation. Computational time is a real bottleneck in this process, with significant impact on the efficiency of the health system when it comes to diagnostic users.
The project will realise a parallel software package to analyse and interpret NMR data with highly advanced techniques and capabilities running on a powerful MPP platform from QSW. The main goal of the development is to adopt software packages as an analysis and verification tool, during the design phase of new NMR data analysis equipment for spectroscopy and imaging. This allows to analyse simulated data derived from the design of novel high-performance systems for the study of biological and synthetic samples, in the chemical and the biomedical fields. A version of the package will then be derived for commercial exploitation as a stand-alone software tool, to be adopted by users of installed NMR systems.
Funding SchemeCSC - Cost-sharing contracts
EC3 7LE London