Document Detail


Support vector machines in DSC-based glioma imaging: suggestions for optimal characterization.
MedLine Citation:
PMID:  20564592     Owner:  NLM     Status:  In-Process    
Abstract/OtherAbstract:
Dynamic susceptibility contrast magnetic resonance perfusion imaging (DSC-MRI) is a useful method to characterize gliomas. Recently, support vector machines (SVMs) have been introduced as means to prospectively characterize new patients based on information from previous patients. Based on features derived from automatically segmented tumor volumes from 101 DSC-MR examinations, four different SVM models were compared. All SVM models achieved high prediction accuracies (>82%) after rebalancing the training data sets to equal amounts of samples per class. Best discrimination was obtained using a SVM model with a radial basis function kernel. A correct prediction of low-grade glioma was obtained at 83% (true positive rate) and for high-grade glioma at 91% (true negative rate) on the independent test data set. In conclusion, the combination of automated tumor segmentation followed by SVM classification is feasible. Thereby, a powerful tool is available to characterize glioma presurgically in patients.
Authors:
Frank G Zöllner; Kyrre E Emblem; Lothar R Schad
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine     Volume:  64     ISSN:  1522-2594     ISO Abbreviation:  Magn Reson Med     Publication Date:  2010 Oct 
Date Detail:
Created Date:  2010-09-27     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  8505245     Medline TA:  Magn Reson Med     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1230-6     Citation Subset:  IM    
Affiliation:
Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany. frank.zoellner@medma.uni-heidelberg.de
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Diffusion-weighted spectroscopy: a novel approach to determine macromolecule resonances in short-ech...
Next Document:  Botulinum neurotoxin-A does not spread to distant muscles after intragastric injection: A double-bli...