Classification of burst and suppression in the neonatal electroencephalogram

• Published in: Journal of neural engineering 2008-12, Vol.5 (4), p.402-410
• Main Authors: Löfhede, J, Löfgren, N, Thordstein, M, Flisberg, A, Kjellmer, I, Lindecrantz, K
• Format: Article
• Language: English
• Subjects: Algorithms; Area Under Curve; Artificial Intelligence; Asphyxia Neonatorum - diagnosis; Asphyxia Neonatorum - physiopathology; burst; classification; Data Interpretation, Statistical; Databases, Factual; EEG; Electroencephalography - classification; Electroencephalography - statistics & numerical data; Fysiologi; Humans; Industrial Biotechnology; Industriell bioteknik; Infant; Infant, Newborn - physiology; Medicinteknik; Models, Statistical; neonatal; neonatal care; Neural Networks (Computer); Other Social Sciences not elsewhere specified; Physiology; Reproducibility of Results; ROC Curve; suppression; Övrig annan samhällsvetenskap
• ISSN: 1741-2552; 1741-2560; 1741-2552
• DOI: 10.1088/1741-2560/5/4/005

Fisher's linear discriminant (FLD), a feed-forward artificial neural network (ANN) and a support vector machine (SVM) were compared with respect to their ability to distinguish bursts from suppressions in electroencephalograms (EEG) displaying a burst-suppression pattern. Five features extracted from the EEG were used as inputs. The study was based on EEG signals from six full-term infants who had suffered from perinatal asphyxia, and the methods have been trained with reference data classified by an experienced electroencephalographer. The results are summarized as the area under the curve (AUC), derived from receiver operating characteristic (ROC) curves for the three methods. Based on this, the SVM performs slightly better than the others. Testing the three methods with combinations of increasing numbers of the five features shows that the SVM handles the increasing amount of information better than the other methods.