Meeting Banner
Abstract #0836

Analysis of 1H MRSI Data of Brain Tumours Using LCModel & Whole Tissue Representations

Felix Raschke1, Franklyn Howe1

1Clinical Sciences, St George's University of London, London, United Kingdom

This study takes forward recently published work using the widespread analysis tool LCModel for the analysis of 2D MRSI data of gliomas. LCModel is designed to estimate individual metabolite proportions by fitting a linear combination of metabolite spectra to an in vivo MR spectrum, but here is used to fit representations of grade II and grade IV tumour spectra and normal white matter. Colormaps and histograms are used to visualize the fitted tissue proportions and allow the clear discrimination of grade II and grade IV gliomas.

Keywords

able abnormal accepting according account acquisition actual additional additionally alternatively apparent appearing apply association automated basis beyond biopsy blue brain cancer changing characteristics classification clear clearly clinical components confirmed containing continuity contribution contributions corrected correlation cram created criteria currently cystic dataset decreasing dept diagnosis diffuse displayed distinguished distorted easy edges encoded england ethics evaluate examples excitation expressed find finding fits fitted fitting flair funded gives grade grades green growth heterogeneity highest histogram histograms improve include indicator individual infiltration infiltrative inspection instead investigations kingdom lactate leading local lowest magnitudes mainly matrix metabolic metabolite might mixed mixture modalities need needed nominal normalized novel output patients peaks planning preparation probe procedures proof proportions protocol published radiologic radiotherapy reaching recently representations representing represents residual return reveal sets share short simulated sorted spectra spectral still studies subjects subsequently surgery survival term tissue tool towards training true types understand valuable variability variation variations visual white widespread zero