Treatment and analysis methods for H-1-MRSI glioma data

Metabolic information measured by proton magnetic resonance spectroscopic imaging (H-1-MRSI) has been approved to be a powerful tool for glioma for distinguishing benign and malignant as well as for grading. However, the raw data of H-1-MRSI are affected by various uncertainties. Therefore, The raw data need to be evaluated and pretreated by denoising processing before the analysis. In the paper, The raw data were evaluated by a data set of criteria. If the data in a voxel does not have enough SNR, then denoising process by Wavelet method was used. After the pretreatment process, the data were analyzed by the Z-score analysis method. The results from the analysis were verified validity by histological conclusion in clinical.Mathematics, AppliedSCI(E)CPCI-S(ISTP)