An automated technique for the quantitative assessment of 3D-MRSI data from patients with glioma

Multiparametric Characterization of Grade 2 Glioma Subtypes Using Magnetic Resonance Spectroscopic, Perfusion, and Diffusion Imaging

“…Median intensities of choline (Cho), creatine (Cre), and N -acetyl aspartate (NAA) were determined from tumor voxels and were normalized by median values of normal voxels in the same patient. The Cho-to-NAA index (CNI) was estimated based on the differences in relative peak heights between tumor and normal tissues [24]. Briefly, an iterative algorithm is used to select a population of voxels that have the spectral features of normal brain regions; then the selected voxels are used as internal controls to quantify the probability of abnormality at each voxel location.…”

Classification of brain tumours using short echo time 1H MR spectra

“…Several studies [5][6][7][8][9][10][11][12][13][14][15][16][17] have already shown progress in automated pattern recognition for brain tumour classification based on MRS data. Several partners from the EU funded INTERPRET project (IST-1999-10310) [18], who provided the data for this study, have already published promising results for classification of brain tumours based on MRS data available within the project [7,12,15,16,[19][20][21][22][23][24].…”

Multi-parametric (ADC/PWI/T2-w) image fusion approach for accurate semi-automatic segmentation of tumorous regions in glioblastoma multiforme

“…These methods, while being advantageous in reducing the expert's labor for manual segmentation, rely on the signal intensity and the boundaries of the image objects; hence, they can only be applied on the anatomical images and are not proper to be applied on the parametric maps like ADC maps or rCBV maps. However, numerous studies have reported detection of brain tumor outside the regions specified by conventional MRI [33], which require physiological MRI techniques, such as PWI and DWI to be identified. Many of the proposed segmentation algorithms are based on classification or clustering approaches, which can handle a pile of information provided by various spectral images.…”