Abstract #1045

Lateralization of Temporal Lobe Epilepsy Using Multimodal MRI, Decision Tree, and Random Forest Methods

Alireza Fallahi¹, Neda Mohammadi-Mobarakeh², Narges Hosseini Tabatabaei³, Mohammad Pooyan⁴, Jafar Mehvari-Habibabadi⁵, Mohammad-Reza Ay², and Mohammad-Reza Nazem-Zadeh²

¹Shahed University, Tehran, Iran (Islamic Republic of), ²Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of), ³Brain and Spinal Cord Injury Research Centre, Neuroscience Institute, Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of), ⁴Biomedical Engineering Department, Shahed University, Tehran, Iran (Islamic Republic of), ⁵dr.mehvari@hotmail.com, Isfahan, Iran (Islamic Republic of)

In this study, a decision making method was developed for determination epileptogenicity in mesial temporal lobe epilepsy (mTLE) patients using different neuroimaging markers including hippocampal volume, and FLAIR (Fluid Attenuated Inversion Recovery) intensity and MD (Mean Diffusivity) value in hippocampus, FA (Fractional anisotropy) in posteroinferior cingulum, and FA in crus of fornix from MRI images of T1, FLAIR, and DTI (diffusion tensor imaging). The aim of this study is to creating an automated classification algorithm using decision tree and random forest methods. Result of applied method detected essential rules for prediction of laterality in individual mTLE patients.

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

Join Here