RESUMO
An important step in the implementation of three-dimensional in vivo proton magnetic resonance spectroscopic imaging ((1)H-MRSI) of the prostate is the placement of spatial saturation pulses around the region of interest (ROI) for the removal of unwanted contaminating signals from peripheral tissue. The present study demonstrates the use of a technique called conformal voxel magnetic resonance spectroscopy (CV-MRS). This method automates the placement, orientation, timing and flip angle of very selective saturation (VSS) pulses around an irregularly-shaped, user-defined ROI. The method employs a user adjustable number of automatically positioned VSS pulses (20 used in the present study) which null the signal from periprostatic lipids while closely conforming the shape of the excitation voxel to the shape of the prostate. A standard endorectal coil in combination with a torso-phased array coil was used for all in vivo prostate studies. Three-dimensional in vivo prostate (1)H-MRSI data were obtained using the proposed semi-automated CV-MRS technique, and compared with a standard point resolved spectroscopy (PRESS) technique at TE = 130 ms using manual placement of saturation pulses. The in vivo prostate (1)H-MRSI data collected from 12 healthy subjects using the CV-MRS method showed significantly reduced lipid contamination throughout the prostate, and reduced baseline distortions. On average there was a 50 ± 17% (range 12% - 68%) reduction in lipids throughout the prostate. A voxel-by-voxel benchmark test of over 850 voxels showed that there were 63% more peaks fitted using the LCModel when using a Cramer-Rao Lower Bound (CRLB) cut-off of 40% when using the optimized conformal voxel technique in comparison to the manual placement approach. The evaluation of this CV-MRS technique has demonstrated the potential for easy automation of the graphical prescription of saturation bands for use in (1)H-MRSI.
Assuntos
Biomarcadores/análise , Diagnóstico por Computador/métodos , Espectroscopia de Ressonância Magnética/métodos , Próstata/anatomia & histologia , Próstata/metabolismo , Adulto , Idoso , Humanos , Masculino , Pessoa de Meia-Idade , Prótons , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Técnica de Subtração , Distribuição TecidualRESUMO
We present measurements of the recoil proton polarization for the d(gamma-->,p-->)n reaction at straight theta(c.m.) = 90 degrees for photon energies up to 2.4 GeV. These are the first data in this reaction for polarization transfer with circularly polarized photons. The induced polarization p(y) vanishes above 1 GeV, contrary to meson-baryon model expectations, in which resonances lead to large polarizations. However, the polarization transfer Cx does not vanish above 1 GeV, inconsistent with hadron helicity conservation. Thus, we show that the scaling behavior observed in the d(gamma,p)n cross sections is not a result of perturbative QCD. These data should provide important tests of new nonperturbative calculations in the intermediate energy regime.