Language lateralization represented by spatiotemporal mapping of magnetoencephalography

Area of Science:

• Neuroscience
• Medical Imaging
• Epileptology

Background:

• Accurate determination of hemispheric language dominance is crucial for epilepsy surgery planning.
• Magnetoencephalography (MEG) offers a non-invasive approach to assess brain function.
• Spatiotemporal source analysis of MEG data is explored for its utility in language lateralization.

Purpose of the Study:

• To evaluate the effectiveness of spatiotemporal source analysis of MEG in determining language laterality.
• To compare the dSPM-amplitude and dSPM-counting methods for language dominance assessment.
• To assess the potential of MEG-based methods to replace or reduce the need for the intracarotid amobarbital procedure (IAP).

Main Methods:

• Thirty-five epilepsy patients performed a semantic word-processing task during MEG recording.
• Dynamic statistical parametric mapping (dSPM) was used to analyze neuromagnetic activity between 250 and 550 ms post-stimulus.
• Laterality indices (LI) were calculated using dSPM-amplitude and dSPM-counting methods, with LI thresholds defining dominance or bilateral representation.
• All patients also underwent the intracarotid amobarbital procedure (IAP) for comparison.

Main Results:

• The dSPM-counting method achieved 91.4% concordance with IAP results for language laterality.
• The dSPM-amplitude method showed only 51.4% concordance with IAP.
• No cases of language laterality opposite to the IAP findings were observed with the dSPM-counting method.
• The dSPM-counting method identified bilateral language representation in 8.6% of patients.

Conclusions:

• Spatiotemporal mapping of language lateralization using the dSPM-counting method is a reliable tool.
• This non-invasive MEG technique demonstrates high accuracy in determining language dominance.
• The dSPM-counting method may significantly reduce the necessity for invasive procedures like the IAP in presurgical epilepsy evaluation, potentially in up to 90% of cases.