University of Twente Student Theses

Login

Improving the nurse response to seizures in the epilepsy monitoring unit with help of EEG-based automatic seizure detection

Jansen Holleboom, L.I.M. (2016) Improving the nurse response to seizures in the epilepsy monitoring unit with help of EEG-based automatic seizure detection.

[img] PDF
2MB
Abstract:In an Epilepsy Monitoring Unit (EMU) it is important that seizures are recognized accurately for diagnostic and safety purposes.In practice sometimes seizures are missed or recognized late. We studied the nurse response in the EMU at SEIN and investigated possible EEG-based improvements. The nurse response and nurse response time was investigated on this data. Secondly, it was investigated if commercially available methods could improve the nurse response. Lastly, we studied a selection of promising features. For this study 205 seizures were included. In the current EMU setting 67.0% of the seizures resulted in a nurse response with a median nurse response time of 32 seconds (p5-p95; 12 - 106 seconds). The commercial software of BESA and AIT detected respectively 38.7% and 66.1% of the seizures without a nurse response. The offline improvement in response time was respectively 25.6 and 18.1 seconds. Lastly, for the features line length and the power over the wavelet coefficients in the majority of the seizures a significant change was observed. In this study sensitivity and latency performances were taken into account, while specificity is also an important feature. Further, it could be discussed what kind of output would be most applicable in an EMU setting.
Item Type:Essay (Master)
Clients:
1989
Faculty:TNW: Science and Technology
Subject:44 medicine
Programme:Technical Medicine MSc (60033)
Link to this item:https://purl.utwente.nl/essays/70490
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page