MOLECULAR SPECTROSCOPY

Author: SANCHEZ GONZALEZ JAVIER.
Year: 2003.
University: POLITÉCNICA DE MADRID [www.upm.es].
Place of defense: ETSI DE TELECOMUNICACION.
Place of preparation: ETSI DE TELECOMUNICACION.

Summary: The application of magnetic resonance spectroscopy (ERM) in clinical studies involved a number of problems in its quantification. This thesis focuses on two of them: the correction of the effects of partial volume and artifacts produced by the reconstruction Cartesian imaging applied to multi-channel spectroscopic imaging. This thesis proposes a new form of analysis that combines the correction of the effects of partial volume with the absolute quantification, using water as a reference signal internal concentrations. This new form of analysis has been successfully evaluated by experiments with phantoms and studies in the frontal lobe of healthy subjects. Moreover, a tool for quantifying MRA studies combining algorithms analyzed in the time domain are better adapted to in vivo studies with the methods needed to correct the effects of partial volume. On the other hand, proposes a new method of reconstruction for spectroscopic imaging techniques learned through multi-channel (SENSE). This reconstruction is based on the properties of the solution norma-mínima to optimize the response function space (SRF) and the role of dispersion point (PSF) of the same. The evaluation was performed by phantoms numerical studies of magnetic resonance imaging studies and image espectroscópica. The results show that this new reconstruction exhibits less pollution in unwanted positions, improving the measurement of this type of study, with respect to reconstruction conventional Cartesian. Finally, the last chapter proposes an algorithm that reduces the computation time required to obtain the solution of norma-mínima. This algorithm is based on the Lanczos method and allows for a sharp acceleration time of calculation with respect to conventional methods.