Line length as a robust method to detect high-activity events: Automated burst detection in premature EEG recordings

• Published in: Clinical neurophysiology Vol. 125; no. 10; pp. 1985 - 1994
• Main Authors: Koolen, Ninah, Jansen, Katrien, Vervisch, Jan, Matic, Vladimir, De Vos, Maarten, Naulaers, Gunnar, Van Huffel, Sabine
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ireland Ltd 01.10.2014
• Subjects: Algorithms; Automated detection; Background EEG; Brain - growth & development; Brain - physiology; Brain maturation; Brain monitoring; Brain Waves - physiology; Bursts; Electroencephalography - methods; Electroencephalography - standards; Female; Humans; Infant; Infant, Newborn; Infant, Premature; Interburst intervals; Line length; Neurology; Polysomnography; Preterm; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Background EEG; Brain monitoring; Bursts; Automated detection; Line length; Preterm; Brain maturation; Interburst intervals
• ISSN: 1388-2457; 1872-8952; 1872-8952
• DOI: 10.1016/j.clinph.2014.02.015

•We present a novel method for automated burst detection in premature EEG recordings, using a single feature Line Length based on both the amplitude and frequency content of the signal.•The accuracy of the burst detection method is within the range of 79.54–89.82%, resulting from both high sensitivity and high specificity and a comparable inter-rater agreement.•The implemented algorithm combines the knowledge of multiple neurologists, thereby objectively assessing premature brain function and reducing the costs for the labor intensive visual analysis. EEG is a valuable tool for evaluation of brain maturation in preterm babies. Preterm EEG constitutes of high voltage burst activities and more suppressed episodes, called interburst intervals (IBIs). Evolution of background characteristics provides information on brain maturation and helps in prediction of neurological outcome. The aim is to develop a method for automated burst detection. Thirteen polysomnography recordings were used, collected at preterm postmenstrual age of 31.4 (26.1–34.4)weeks. We developed a burst detection algorithm based on the feature line length and compared it with manual scorings of clinical experts and other published methods. The line length–based algorithm is robust (84.27% accuracy, 84.00% sensitivity, 85.70% specificity). It is not critically dependent on the number of measurement channels, because two channels still provide 82% accuracy. Furthermore, it approximates well clinically relevant features, such as median IBI duration 5.45 (4.00–7.11)s, maximum IBI duration 14.02 (8.73–18.80)s and burst percentage 48.89 (35.45–60.12)%, with a median deviation of respectively 0.65s, 1.96s and 6.55%. Automated assessment of long-term preterm EEG is possible and its use will optimize EEG interpretation in the NICU. This study takes a first step towards fully automatic analysis of the preterm brain.