A seizure warning system for long-term epilepsy monitoring

We developed a system to provide a warning early in the development of a seizure with a reasonably low false alarm rate.Such a warning will improve the close observation of seizures and interaction between observers and patients early in the seizure, even in seizures having no obvious clinical manifestation. The system relies on the availability of the EEG recording of one sample seizure, which is used as a template for subsequent detection. We evaluated the performance in 24 seizure types from scalp and intracerebral recordings. It yielded a 100% detection rate and a false alarm rate averaging one false detection every 5 hours. The warning signal was given, on average, 9.6 seconds after EEG seizure onset. The system will be useful in improving the clinical observation of seizures and may allow ictal SPECT scans to be more widely performed. NEUROLOGY 1995;45: 2250-2254

[1]  J. Gotman.  Automatic recognition of epileptic seizures in the EEG. , 1982, Electroencephalography and clinical neurophysiology.

[2]  Josef Kittler,et al.  Pattern recognition : a statistical approach , 1982 .

[3]  SINGLE PHOTON EMISSION TOMOGRAPHY , 1982, The Lancet.

[4]  J. Gotman.  Automatic seizure detection: improvements and evaluation. , 1990, Electroencephalography and clinical neurophysiology.

[5]  F. Pierelli,et al.  Long-term EEG-video-audio monitoring: computer detection of focal EEG seizure patterns. , 1992, Electroencephalography and clinical neurophysiology.

[6]  A. Liu,et al.  Detection of neonatal seizures through computerized EEG analysis. , 1992, Electroencephalography and clinical neurophysiology.

[7]  J. Gotman,et al.  Improvement in seizure detection performance by automatic adaptation to the EEG of each patient. , 1993, Electroencephalography and clinical neurophysiology.

[8]  C. Jack,et al.  Determination of 10-20 system electrode locations using magnetic resonance image scanning with markers. , 1993, Electroencephalography and clinical neurophysiology.

[9]  J. Gotman,et al.  A patient-specific algorithm for the detection of seizure onset in long-term EEG monitoring: possible use as a warning device , 1997, IEEE Transactions on Biomedical Engineering.