Ιδρυματικό Αποθετήριο
Πολυτεχνείο Κρήτης
Πολυτεχνείο Κρήτης
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/892FCE93-ECD2-4A47-9CC0-04DCA4E11306 | ||
| Έτος | 2010 | ||
| Τύπος | Πλήρης Δημοσίευση σε Συνέδριο | ||
| Άδεια Χρήσης |
|
||
| Βιβλιογραφική Αναφορά | M.G. Kounelakis, M.E. Zervakis, G.J. Postma, L.M.C. Buydens, G.C. Giakos, C. Narayan, S. Marotta, D. Natarajamani, X. Kotsiakis ,"Targeting brain gliomas energy metabolism for classification purposes ," in 2010 IEEE Intern. Conf. on Imaging Syst.and Techniques (IST) ,pp.36-40.doi:10.1109/IST.2010.5548526 https://doi.org/10.1109/IST.2010.5548526 | ||
| Εμφανίζεται στις Συλλογές |
The aim of this study is to reveal the discriminative potential of energy related metabolites in brain gliomas classification. The proposed analysis considers two aspects, the statistical and biological verification of metabolites' effects. In particular, Magnetic Resonance Spectroscopic Imaging (MRSI) is first employed for the statistical evaluation of metabolites. Five of the identified significant metabolites, namely glucose, pyruvate, lactate, alanine and lipids, are involved in the energy production process, necessary for the survival of the cell. In the second stage of analysis, we consider specific metabolic pathways like glycolysis, lactate fermentation, citric acid cycle and lipogenesis in order to evaluate the role of these metabolites in the energy production process. The results of the proposed double process have shown that these five metabolites are capable to discriminate the three types of gliomas provided. Low grade glioma (GR2) can be discriminated from intermediate grade ones (GR3) with accuracy of 86%. Intermediate grade gliomas versus high grade ones (GR4) with accuracy of 98% and finally low grade versus high grade with accuracy of 100%. Accuracy was extensively evaluated with the area under the ROC (AUROC) global metric.
Copyright © DIAS 2013
