| |
Seizures occur in 1 to 3.5/1000 births and are a common sign of neurological dysfunction in neonates. In a strategy of neuroprotection of the newborn, effective detection of these seizures is needed in order to assess their potential additional damage and establish timely treatment. The detection of seizures is usually based on clinical observation in conjunction with visual assessment of the EEG. In neonates, the clinical seizures are often subtle and may be missed without constant supervision. Furthermore, most seizures tend to be subclinical, implying that they can be detected only by EEG monitoring. Combined with the fact that EEG analysis requires particular skills which are not always present around the clock in the Neonatal Intensive Care Unit (NICU), this means that many seizures are missed. An automated system that reliably detects neonatal seizures would be of significant value in the NICU. Therefore we developed an automated neonatal seizure detection algorithm which mimics the human observer. To lower the false positive rate of the detector, an artifact removal strategy, based on blind source separation (BSS), is implemented. After a seizure is detected, the spatial distribution of the seizure is extracted using PARAFAC decomposition and is visualized to the user. These three algorithms can be combined into a single seizure monitoring system at the bedside. |
|
