Wylezinska-Arridge, Marzena Malgorzata; (1997) Localized phosphorus spectroscopy in vivo: Quantitation of metabolite concentrations. Doctoral thesis (Ph.D), UCL (University College London).

Text Localized_phosphorus_spectrosc.pdf Download (10MB)

Abstract

This project was dedicated to the investigation of the factors that may affect absolute quantitation in localized 31P MRS and if possible to the improvement of the accuracy of both localization and quantification. Three aspects have been looked at: 1) the acquisition /localization technique used; 2) the strategy used for conversion of signal amplitude/peak areas into concentrations; and 3) methods for MRS signal processing and analysis. With respect to the first aspect, image selected in vivo spectroscopy (ISIS) and point resolved spectroscopy (PRESS), were considered. Aspects of ISIS localization, including relaxation effects during inversion and excitation adiabatic pulses, and uniformity of spin excitation across the "in vivo" 31P spectral range, were investigated using simulation. In order to reduce the chemical shift displacement error in ISIS, a new adiabatic pulse for spin inversion, has been designed and experimentally verified. For PRESS, the performance of the selective 90[degrees] and 180[degrees] pulses was investigated experimentally and using simulations. The consequences of nonideal flip angles on T1 measurements based on two PRESS experiments were analyzed. Effects of amplitude and phase modulation of the ATP signal during the PRESS sequence were analyzed using product-operator formalism for an AMX system. A tissue substitute material, with known metabolite concentrations and simulating the 31P spectrum obtained from neonatal brain, has been developed for testing quantitation accuracy. The manufacture, physical properties and chemical stability of a material has been presented. The following calibration protocols have been experimentally verified: use of water as an internal concentration reference (ICR), and use of a standard phantom as an external concentration reference (ECR). A modified ECR protocol using the tissue substitute material as a reference, has been suggested to deal with problems related to off-resonance effects. This protocol has been applied to "in vivo" 31P studies of calf muscle and neonatal brain. Results obtained in human calf muscle agree with previously published data. For data analysis the time domain techniques VARPRO, LPSVD, and HLSVD were considered and compared with X2 minimization based curve fitting in the frequency domain. Concentrations obtained with VARPRO and curve fitting in the frequency domain show no significant differences provided was that all calibration measurement were analyzed with the same analysis technique.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item