Multiparametric MRI Analysis of Glioblastoma Multiforme Tissues Using Multi-Class Support Vector Machines

Keywords

accuracy achieve advantages agrees alone although aside audience basic biomedical bold breathing capable class classification classified classifier classifying clinical collected complex condition confirmed consisting contrast corrected degree delta detection determine diffusion disease early edema either engineering exaggerated expected feature features focus future giving help hospital hypoxia improve improved increasing input inspection introduced john labels leakage learning longitudinal losses machine machines maps matrix medical misclassification model models necrosis need networks obvious offer optimizing outperforms overlaid oxygenation patient patients performance perfusion physicist physics pool post potential predicting prove radiologist radiology recurrence recurrent registered represented roughly seek sensitivity session simple specificity subjects substantial subtraction suffered support target tissue tissues trained truly tumor tumors types various vector visual yellow