Predicting MGMT Methylation Status of Glioblastomas from MRI Texture Ilya Levner 1,2,3 , Sylvia Drabycz 4 , Gloria Roldan 5,6,7,8 , Paula De Robles 5,6,7,8 , J. Gregory Cairncross 5,6,7,8 , and Ross Mitchell 1,2,3,6,7,8 1 Department of Radiology, University of Calgary, Alberta, Canada 2 Southern Alberta Cancer Research Institute, University of Calgary, Alberta, Canada 3 Alberta Ingenuity Center for Machine Learning, Canada 4 Department of Electrical and Computer Engineering, University of Calgary, Alberta, Canada 5 Department of Oncology, Tom Baker Cancer Centre, Alberta Cancer Board, Canada 6 Hotchkiss Brain Institute, Calgary, Alberta, Canada 7 Department of Clinical Neurosciences, University of Calgary, Alberta Canada 8 Clark Smith Brain Tumor Centre, Southern Alberta Cancer Research Institute, Canada Abstract. In glioblastoma (GBM), promoter methylation of the DNA repair gene MGMT is associated with benefit from chemotherapy. Be- cause MGMT promoter methylation status can not be determined in all cases, a surrogate for the methylation status would be a useful clinical tool. Correlation between methylation status and magnetic resonance imaging features has been reported suggesting that non-invasive MGMT promoter methylation status detection is possible. In this work, a ret- rospective analysis of T2, FLAIR and T1-post contrast MR images in patients with newly diagnosed GBM is performed using L1-regularized neural networks. Tumor texture, assessed quantitatively was utilized for predicting the MGMT promoter methylation status of a GBM in 59 patients. The texture features were extracted using a space-frequency texture analysis based on the S-transform and utilized by a neural net- work to predict the methylation status of a GBM. Blinded classification of MGMT promoter methylation status reached an average accuracy of 87.7%, indicating that the proposed technique is accurate enough for clinical use. 1 Introduction Glioblastoma multiforme (GBM) is the most common primary brain tumor in adults. Standard treatment now includes a DNA alkylating agent, Temozolo- mide (TMZ), which is the only known chemotherapeutic that prolongs survival [1]. Interestingly, the effectiveness of TMZ may be predictable; via a test for methylation of the O6-methylguanine-DNA methyltransferase (MGMT) gene G.-Z. Yang et al. (Eds.): MICCAI 2009, Part II, LNCS 5762, pp. 522–530, 2009. c  Springer-Verlag Berlin Heidelberg 2009