Meeting Banner
Abstract #1964

One-Class Classifier for Accurate Brain Tissue Classification from Noisy 1H-MRS Spectra

Keyvan Ghassemi 1,2 , Mohammadreza Khanmohammadi Khorami 1 , and Hamidreza Saligheh Rad 2,3

1 Chemistry Department, Faculty of Science, Imam Khomeini International University, Qazvin, Iran, 2 Quantitative MR Imaging and Spectroscopy Group, Research Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences, Tehran, Iran, 3 Department of Medical Physics and Biomedical Engineering, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran

Low signal to noise ratio (SNR), baseline distortions, large line-widths and asymmetric line-shapes caused by poor shimming, as well as contaminations caused by significant chemical shift displacement effects produce complicated MRS signals. Totally 139 spectra from healthy and tomure glial brains 10 healthy cases,11 grade II, 6 grade III, as well as 9 grade IV brain gliomas were collected. SIMCA was used by application of PCA in common rule and by using of the NMF. Results of robust SIMCA showed significant modification in percentage of correct classified samples after application of NMF for better decomposition of noisy measurements.

This abstract and the presentation materials are available to members only; a login is required.

Join Here