Robustness of implantable algorithms to detect epileptiform activity in the presence of broad-spectrum background noise.

Detection of epileptiform activity is of interest for responsive stimulation and diagnostic or monitoring devices in epilepsy; some implantable systems use low-computational-complexity algorithms such as line length trending and half-wave detection. Broadband noise was added to recorded electrocorticographic signals in order to model the potential impact of factors such as electrode-tissue interface properties and distance from the epileptic focus on these detection tools. Simulation demonstrated that half-wave and line length tools can yield consistent results in the presence of moderate amounts of noise.

Automated seizure onset detection for accurate onset time determination in intracranial EEG.

Objective: A novel algorithm for automated seizure onset detection is presented. The method allows for precise identification of electrographic seizure onset times within large databases of electrographic data.

Methods: The patient-specific algorithm extracts salient spectral and temporal features in five frequency bands within a sliding window of an electrographic recording.