Authors:
A. Temko
;
I. Korotchikova
;
W. Marnane
;
G. Lightbody
and
G. Boylan
Affiliation:
University College Cork, Ireland
Keyword(s):
Neonatal, Seizure, Detection, Automated, Energy, Normalization, Support vector machines, Healthy patients, False detections per hour.
Related
Ontology
Subjects/Areas/Topics:
Applications and Services
;
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computer Vision, Visualization and Computer Graphics
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Medical Image Detection, Acquisition, Analysis and Processing
;
Methodologies and Methods
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Soft Computing
Abstract:
Seizures in newborn babies are commonly caused by problems such as lack of oxygen, haemorrhage, meningitis, infection and strokes. The aim of an automated neonatal seizure detection system is to assist clinical staff in a neonatal intensive care unit to interpret the EEG. In this work, the automated neonatal seizure detection system is validated on a set of healthy patients and its performance is compared to the performance obtained on sick patients reported previously. The histogram-based energy normalization technique is designed and applied to EEG signals from healthy patients to cope with montage mismatch. The results on healthy babies compares favourably to those obtained on sick babies. Several useful observations are made which were not possible to obtain by testing on sick babies only such as a practically useful range of probabilistic thresholds, minimum detection duration restriction, and an influence of the database statistics on the system performance.