Abstract #4238

Prediction of IDH Mutation Status of Diffuse-Gliomas Based on Short-Echo Time Magnetic Resonance Spectroscopy at 3T

Esin Ozturk-Isik¹, Sevim Cengiz¹, Koray Ozduman^2,3, Alpay Ozcan^3,4, Cengiz Yakicier⁵, M. Necmettin Pamir^2,3, and Alp Dincer^3,6

¹Institute of Biomedical Engineering, Bogazici University, Istanbul, Turkey, ²Department of Neurosurgery, Acibadem University, Istanbul, Turkey, ³Neuroradiology Research Center, Acibadem University, Istanbul, Turkey, ⁴Department of Medical Engineering, Acibadem University, Istanbul, Turkey, ⁵Department of Molecular Biology and Genetics, Acibadem University, Istanbul, Turkey, ⁶Department of Radiology, Acibadem University, Istanbul, Turkey

The tumor biology of diffuse-gliomas is best reflected by their molecular profile. Isocitrate dehydrogenase (IDH) mutation status has the strongest correlation to treatment response and patient survival among all molecular markers. The aim of this study is to predict IDH mutation status of gliomas based on short-echo time MR spectroscopic (MRS) biomarkers by using machine learning algorithms at 3T. Our results indicated that MRS based biomarkers were able to discriminate between IDH-mutant and IDH-wild type patients with up to 91% sensitivity, %78 specificity, and 86.9% accuracy by using an ensemble of bootstrap-aggregated decision trees classification with a three-fold cross validation.

This abstract and the presentation materials are available to members only; a login is required.

Join Here