Identification of significant metabolic markers from MRSI data for brain cancer classification

Investigation of the significance of metabolites peak area ratios derived from brain magnetic resonance spectroscopic imaging (MRSI) spectra, in brain tumors classification, has been applied. Results have shown that in most binary classifications using SVM and LSSVM classifiers, the accuracy achieved was greater than 0.90 AUC except the case of Gliomas grade 2 vs Gliomas grade 3 where 0.84 AUC was recorded due to the great heterogeneity of these two types of tumor. The minimum but also biologically significant set of features (markers), where maximum AUCs recorded, was derived. Ratios of N-acetyl-aspartate, choline, creatine and lipids metabolites found to play the most crucial role in brain tumors discrimination. The biological importance of these markers was also verified by literature. Finally the influence of four magnetic resonance image (MRI) intensities on the classification process was also measured. It was found that MRI data do not improve significantly the classification accuracies.

[1]  Arend Heerschap,et al.  A chemometric approach for brain tumor classification using magnetic resonance imaging and spectroscopy. , 2003, Analytical chemistry.

[2]  Leif E. Peterson,et al.  Logistic Ensembles for Random Spherical Linear Oracles , 2007, ICMLA 2007.

[3]  Irina Mader,et al.  Assignment of glial brain tumors in humans byin vivo1H-magnetic resonance spectroscopy and multidimensional metabolic classification , 1997, Magnetic Resonance Materials in Physics, Biology and Medicine.

[4]  Theodoros N. Arvanitis,et al.  The use of short-echo-time 1H MRS for childhood cerebellar tumours prior to histopathological diagnosis , 2007, Pediatric Radiology.

[5]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[6]  A Kumar,et al.  Role of in vivo proton MR spectroscopy in the evaluation of adult brain lesions: our preliminary experience. , 2003, Neurology India.

[7]  Michalis E. Blazadonakis,et al.  Polynomial and RBF Kernels as Marker Selection Tools-A Breast Cancer Case Study , 2007, Sixth International Conference on Machine Learning and Applications (ICMLA 2007).

[8]  Franklyn A Howe,et al.  1H MR spectroscopy of brain tumours and masses , 2003, NMR in biomedicine.

[9]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[10]  Mathias Hoehn,et al.  Metabolic differences between primary and recurrent human brain tumors: a 1H NMR spectroscopic investigation , 2005, NMR in biomedicine.

[11]  Vladimir Cherkassky,et al.  The Nature Of Statistical Learning Theory , 1997, IEEE Trans. Neural Networks.

[12]  F. Howe,et al.  Differentiation of metastases from high‐grade gliomas using short echo time 1H spectroscopy , 2004, Journal of magnetic resonance imaging : JMRI.

[13]  D A Porter,et al.  Magnetic resonance spectroscopy in vivo. , 1988, Journal of biomedical engineering.

[14]  Z. Wu,et al.  In vivo single-voxel proton MR spectroscopy in the differentiation of high-grade gliomas and solitary metastases. , 2004, Clinical radiology.

[15]  B. Ross,et al.  Clinical Applications of Magnetic Resonance Spectroscopy , 1997, Epilepsia.