Abstracts

Lateralized neuromagnetic inferior frontal gyrus activation during auditory word recognition: correlation with Wada test results

Abstract number : 267
Submission category : 3. Neurophysiology / 3D. MEG
Year : 2020
Submission ID : 2422613
Source : www.aesnet.org
Presentation date : 12/6/2020 12:00:00 PM
Published date : Nov 21, 2020, 02:24 AM

Authors :
Hyun Freeman, University of Alabama, Birmingham; Jeffery Killen - University of Alabama, Birmingham; Roy Martin - University of Alabama at Birmingham; Ismail Mohamed - University of Alabama, Birmingham;


Rationale:
Auditory word recognition recruits a bilateral neural network. The most consistent activation tends to occur in the superior temporal gyri where initial processing of speech sounds occurs. Early activation of the left inferior frontal gyrus (IFG) was observed in evoked potential studies and is thought to represent early phonemic processing. Magnetoencephalography has been used for presurgical language lateralization often using an auditory word recognition task. The most commonly used analysis technique, the equivalent current dipole, estimates number of dipolar sources in the perisylvian core and then calculates a lateralization index (LI) as a representation of language laterality. As IFG activation during auditory word recognition is of smaller magnetic magnitude, dipole modelling of inferior frontal activity is often unreliable.The goal of this study is to evaluate whether a new computational approach based on oscillatory brain response, maximum entropy of the mean (wMEM) can provide a better representation of inferior frontal gyrus activation and whether lateralized inferior frontal activity would correlate with Wada test and surgical mapping results
Method:
MEG data was collected on 16 patients between the ages of 13-62 years old with intractable temporal lobe epilepsy or temporal lobe space occupying lesions evaluated for surgery. Each patient completed two runs of an auditory word recognition task. Individual subject’s brain was segmented using freesurfer software. Time-frequency representation of the averaged data was applied to determine the dominant frequency in the IFG. We implemented a wavelet based maximum entropy of the mean (wMEM) dedicated to performing source localization of oscillatory patterns in the time-frequency domain. Source maps in the 5-15 Hz band, 200-500 milliseconds were calculated then LI was calculated based on activation scores within scouts  over the pars orbicularis and pars triangularis of the right and left inferior frontal gyri and compared to Wada test results or if not performed to cortical mapping performed during epilepsy surgery.
Results:
MRI was normal in 8 patients, Brain tumors in 5 patients, cortical dysplasia in 2 and hippocampal sclerosis in one patient. Wada test was performed on 9 patients and invasive monitoring was performed on 12 patients using subdural electrodes in 8 and SEEG in four patients. Four patients with brain tumors underwent direct surgery with intraoperative language mapping. IFG oscillatory activity were left lateralized in 12 patients and bilateral in four. LI was concordant with Wada test results in seven patients. Two patients had bilateral IFG LI but lateralized Wada test results. IFG LI was concordant with intraoperative or extracoperative invasive cortical mapping in the remaining seven patients.
Conclusion:
Patients with temporal lobe epilepsy tend to show more robust oscillatory IFG activation during auditory word recognition. This response is lateralized and correlated with invasive cortical mapping in this small group of patients.
Funding:
:The study is funded through KPRI grant and NSF EPSCor
Neurophysiology