Felix Raschke1, Elies Fuster-Garcia2,3, Kirstie Suzanne Opstad1, Franklyn Arron Howe1
1Division of Clinical Science, St Georges University of London, London, United Kingdom; 2IBIME & ITACA, Universidad Politcnica de Valencia, Valencia, Spain; 3Universitat Internacional Valenciana, Valencia, Spain
This study presents a novel method for the direct classification of single voxel 1H MR spectra of brain tumours using the widespread analysis tool LCModel. 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 complete tumour spectra. Classification according to the highest estimated tumour proportion in a test set of 46 spectra comprising high grade gliomas, low grade gliomas and meningiomas, LCModel gives a classification accuracy of 93% compared to 95% by the INTERPRET Decision Support System.