Seizure lateralization in scalp EEG using Hjorth parameters.

Area of Science:

• Neuroscience
• Medical Technology

Background:

• Temporal lobe epilepsy (TLE) is a common neurological disorder.
• Accurate lateralization of seizure onset is crucial for diagnosis and treatment planning.
• Current methods for seizure lateralization can be time-consuming and subjective.

Purpose of the Study:

• To develop and evaluate a novel semi-automatic method for lateralizing temporal lobe seizures.
• To utilize raw scalp electroencephalogram (EEG) signals for seizure lateralization.
• To provide an objective and efficient tool for neurophysiologists.

Main Methods:

• Employed Hjorth parameters (quadratic mean and dominant frequency) from scalp EEG.
• Segmented EEG data around seizure onset to isolate initial seizure activity.
• Characterized seizures in a frequency-amplitude plane and applied partitioning criteria for lateralization.

Main Results:

• The method achieved a high correct lateralization rate of 96% with the best criterion.
• An alternative criterion yielded 87% correct lateralization without false positives.
• The procedure demonstrated robustness to artifacts and simplicity in parameter setting.

Conclusions:

• The developed semi-automatic method effectively lateralizes temporal lobe seizures.
• The technique is user-friendly, interpretable, and reliable.
• This approach can assist neurophysiologists in interpreting EEG data and diagnosing epilepsy.