Texture analysis applied to brain MRI to classify low and high grade gliomas

Gliomas can be classified as either low grade glioma (LGG) or high grade glioma (HGG). Standard diagnosis is based on histopathological tests obtained from a surgical resection or a stereotactic biopsy. Due to their heterogeneity, these tumors can be misclassified. Therefore, there is a need to develop non-invasive and automatic methods that could help specialists with their correct classification. The aim of this work was to develop a computational classification method which distinguished LGGs from HGGs, based on texture analysis of magnetic resonance images (MRI). The model reported was based on a simple methodology and proved to be useful for the classification of gliomas.