Ninah Koolen1, Katrien Jansen2, Jan Vervisch2, Vladimir Matic3, Maarten De Vos4, Gunnar Naulaers5, Sabine Van Huffel3. 1. Department of Electrical Engineering (ESAT), Division SCD, KU Leuven, Leuven, Belgium; iMinds-KU Leuven Future Health Department, Leuven, Belgium. Electronic address: ninah.koolen@esat.kuleuven.be. 2. Department of Pediatrics, University Hospital Gasthuisberg, Leuven, Belgium. 3. Department of Electrical Engineering (ESAT), Division SCD, KU Leuven, Leuven, Belgium; iMinds-KU Leuven Future Health Department, Leuven, Belgium. 4. Cluster of Excellence "Hearing4all" & Methods in Neurocognitive Psychology, University of Oldenburg, Oldenburg, Germany; Department of Electrical Engineering (ESAT), Division SCD, KU Leuven, Leuven, Belgium. 5. Neonatal Intensive Care Unit, University Hospital Gasthuisberg, Leuven, Belgium.
Abstract
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.
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.
Authors: Anneleen Dereymaeker; Kirubin Pillay; Jan Vervisch; Maarten De Vos; Sabine Van Huffel; Katrien Jansen; Gunnar Naulaers Journal: Early Hum Dev Date: 2017-07-12 Impact factor: 2.079
Authors: Seungha Lee; Xuelong Zhao; Kathryn A Davis; Alexis A Topjian; Brian Litt; Nicholas S Abend Journal: Neurology Date: 2019-04-10 Impact factor: 9.910
Authors: Franz Fürbass; Johannes Herta; Johannes Koren; M Brandon Westover; Manfred M Hartmann; Andreas Gruber; Christoph Baumgartner; Tilmann Kluge Journal: Clin Neurophysiol Date: 2016-02-09 Impact factor: 3.708
Authors: N Koolen; A Dereymaeker; O Räsänen; K Jansen; J Vervisch; V Matic; G Naulaers; M De Vos; S Van Huffel; S Vanhatalo Journal: Neuroscience Date: 2016-02-11 Impact factor: 3.590
Authors: John M O'Toole; Geraldine B Boylan; Rhodri O Lloyd; Robert M Goulding; Sampsa Vanhatalo; Nathan J Stevenson Journal: Med Eng Phys Date: 2017-04-18 Impact factor: 2.242