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

OBJECTIVE 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. METHODS 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. RESULTS 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%. CONCLUSION Automated assessment of long-term preterm EEG is possible and its use will optimize EEG interpretation in the NICU. SIGNIFICANCE This study takes a first step towards fully automatic analysis of the preterm brain.

[1]  B. Bourgeois,et al.  Prognostic value of neonatal discontinuous EEG. , 2002, Pediatric neurology.

[2]  M. De Vos,et al.  Validation of a new automated neonatal seizure detection system: A clinician’s perspective , 2011, Clinical Neurophysiology.

[3]  M. Siegel,et al.  Fanconi's anemia, type A presenting as VACTERL association with atresia right external auditory canal , 2010, Journal of Perinatology.

[4]  M. Vecchierini,et al.  Normal EEG of premature infants born between 24 and 30 weeks gestational age: Terminology, definitions and maturation aspects , 2007, Neurophysiologie Clinique/Clinical Neurophysiology.

[5]  A. Flisberg,et al.  Automatic classification of background EEG activity in healthy and sick neonates , 2010, Journal of neural engineering.

[6]  Ingmar Rosén,et al.  Electroencephalography and brain damage in preterm infants. , 2005, Early human development.

[7]  A. Furby,et al.  Prognostic value of EEG in very premature newborns , 2011, Archives of Disease in Childhood: Fetal and Neonatal Edition.

[8]  L. Wong,et al.  Time-frequency evaluation of segmentation methods for neonatal EEG signals , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[9]  Lawrence J. Hirsch,et al.  Interrater Reliability of ICU EEG Research Terminology , 2012, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[10]  Sampsa Vanhatalo,et al.  Optimization of an NLEO-based algorithm for automated detection of spontaneous activity transients in early preterm EEG , 2010, Physiological measurement.

[11]  Tapio Seppänen,et al.  Automatic Analysis and Monitoring of Burst Suppression in Anesthesia , 2002, Journal of Clinical Monitoring and Computing.

[12]  M. De Vos,et al.  Automated artifact removal as preprocessing refines neonatal seizure detection , 2011, Clinical Neurophysiology.

[13]  Thomas J. Davidson,et al.  Closed-loop optogenetic control of thalamus as a new tool to interrupt seizures after cortical injury , 2012, Nature Neuroscience.

[14]  Perumpillichira J. Cherian,et al.  Technical standards for recording and interpretation of neonatal electroencephalogram in clinical practice , 2009, Annals of Indian Academy of Neurology.

[15]  A. Okumura,et al.  Background electroencephalographic (EEG) activities of very preterm infants born at less than 27 weeks gestation: a study on the degree of continuity , 2001, Archives of disease in childhood. Fetal and neonatal edition.

[16]  Brian Litt,et al.  Line length: an efficient feature for seizure onset detection , 2001, 2001 Conference Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[17]  K. Kaila,et al.  Development of neonatal EEG activity: from phenomenology to physiology. , 2006, Seminars in fetal & neonatal medicine.

[18]  G. Lightbody,et al.  EEG-based neonatal seizure detection with Support Vector Machines , 2011, Clinical Neurophysiology.

[19]  Hiroyuki Kidokoro,et al.  EEG for Predicting Early Neurodevelopment in Preterm Infants: An Observational Cohort Study , 2012, Pediatrics.

[20]  I. Rosén,et al.  Electroencephalography and brain injury in preterm infants , 2005 .

[21]  Ingmar Rosén,et al.  Amplitude-integrated EEG Classification and Interpretation in Preterm and Term Infants , 2006 .

[22]  P. Agostino Accardo,et al.  Use of the fractal dimension for the analysis of electroencephalographic time series , 1997, Biological Cybernetics.

[23]  S. Bambang Oetomo,et al.  Quantitative analysis of maturational changes in EEG background activity in very preterm infants with a normal neurodevelopment at 1 year of age. , 2010, Early human development.

[24]  A. Mathur,et al.  Using amplitude-integrated EEG in neonatal intensive care , 2010, Journal of Perinatology.

[25]  Carola van Pul,et al.  Automatic burst detection for the EEG of the preterm infant. , 2011, Physiological measurement.

[26]  José Ramón Castro Conde,et al.  Extrauterine life duration and ontogenic EEG parameters in preterm newborns with and without major ultrasound brain lesions , 2005, Clinical Neurophysiology.

[27]  E. Walls-Esquivel,et al.  Electroencephalography in premature and full-term infants. Developmental features and glossary , 2010, Neurophysiologie Clinique/Clinical Neurophysiology.

[28]  S. Huffel,et al.  Neonatal seizure localization using PARAFAC decomposition , 2009, Clinical Neurophysiology.

[29]  G Cioni,et al.  Background EEG activity in preterm infants: correlation of outcome with selected maturational features. , 1994, Electroencephalography and clinical neurophysiology.

[30]  Sabine Van Huffel,et al.  Automatic Burst Detection based on Line Length in the Premature EEG , 2013, BIOSIGNALS.

[31]  J. Volpe,et al.  Neurologic outcome of prematurity. , 1998, Archives of neurology.

[32]  M. André,et al.  Normal EEG in very premature infants: reference criteria , 2000, Clinical Neurophysiology.

[33]  Sampsa Vanhatalo,et al.  Detection of ‘EEG bursts’ in the early preterm EEG: Visual vs. automated detection , 2010, Clinical Neurophysiology.

[34]  Lojini Logesparan,et al.  Optimal features for online seizure detection , 2012, Medical & Biological Engineering & Computing.

[35]  Assuring Healthy Outcomes,et al.  Preterm Birth : Causes , Consequences , and Prevention , 2005 .

[36]  R. Clancy,et al.  Neurologic outcome after electroencephalographically proven neonatal seizures. , 1991, Pediatrics.

[37]  Vineta Fellman,et al.  Early single-channel aEEG/EEG predicts outcome in very preterm infants , 2012, Acta paediatrica.