Display options
Cite
Martin JR, Gabriel P, Gold J, et al. Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG. J Clin Neurophysiol. 2020;doi: 10.1097/WNP.0000000000000767.
Martin, J. R., Gabriel, P., Gold, J., Haas, R., Davis, S., Gonda, D., Sharpe, C., Wilson, S., Nierenberg, N., Scheuer, M., & Wang, S. (2020). Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG. Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society, . https://doi.org/10.1097/WNP.0000000000000767
Martin, Joel R, et al. "Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG." Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society vol. (2020). doi: https://doi.org/10.1097/WNP.0000000000000767
Martin JR, Gabriel P, Gold J, Haas R, Davis S, Gonda D, Sharpe C, Wilson S, Nierenberg N, Scheuer M, Wang S. Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG. J Clin Neurophysiol. 2020 Aug 17; doi: 10.1097/WNP.0000000000000767. Epub 2020 Aug 17. PMID: 32810002; PMCID: PMC7887141.
Copy Download .nbib Format: NLM
Share it on

J Clin Neurophysiol. 2020 Aug 17; doi: 10.1097/WNP.0000000000000767. Epub 2020 Aug 17.

Optical Flow Estimation Improves Automated Seizure Detection in Neonatal EEG.

Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society

Joel R Martin, Paolo Gabriel, Jeffrey Gold, Richard Haas, Sue Davis, David Gonda, Cia Sharpe, Scott Wilson, Nicolas Nierenberg, Mark Scheuer, Sonya Wang

Affiliations

  1. Departments of Electrical Engineering.
  2. Neurosciences.
  3. Pediatrics, University of California, San Diego, California, U.S.A.
  4. Auckland District Health Board, Auckland, New Zealand.
  5. Department of Surgery, University of California, San Diego, California, U.S.A.
  6. Persyst, Solana Beach, California, U.S.A.
  7. Department of Neurology, University of Minnesota, Minneapolis, Minnesota, U.S.A.

PMID: 32810002 PMCID: PMC7887141 DOI: 10.1097/WNP.0000000000000767

Abstract

INTRODUCTION: Existing automated seizure detection algorithms report sensitivities between 43% and 77% and specificities between 56% and 90%. The algorithms suffer from false alarms when applied to neonatal EEG because of the high degree of nurse handling and rhythmic patting used to soothe neonates. Computer vision technology that quantifies movement in real time could distinguish artifactual motion and improve automated neonatal seizure detection algorithms.

METHODS: The authors used video EEG recordings from 43 neonates undergoing monitoring for seizures as part of the NEOLEV2 clinical trial. The Persyst neonatal automated seizure detection algorithm ran in real time during study EEG acquisitions. Computer vision algorithms were applied to extract detailed accounts of artifactual movement of the neonate or people near the neonate though dense optical flow estimation.

RESULTS: Using the methods mentioned above, 197 periods of patting activity were identified and quantified, of which 45 generated false-positive automated seizure detection events. A binary patting detection algorithm was trained with a subset of 470 event videos. This supervised detection algorithm was applied to a testing subset of 187 event videos with 8 false-positive events, which resulted in a 24% reduction in false-positive automated seizure detections and a 50% reduction in false-positive events caused by neonatal care patting, while maintaining 11 of 12 true-positive seizure detection events.

CONCLUSIONS: This work presents a novel approach to improving automated seizure detection algorithms used during neonatal video EEG monitoring. This artifact detection mechanism can improve the ability of a seizure detector algorithm to distinguish between artifact and true seizure activity.

References

  1. Seizure. 2016 Aug;40:88-101 - PubMed
  2. Physiol Meas. 2010 Jul;31(7):1047-64 - PubMed
  3. J Neural Eng. 2012 Aug;9(4):046002 - PubMed
  4. J Clin Neurophysiol. 2009 Aug;26(4):218-26 - PubMed
  5. Curr Opin Pediatr. 2014 Dec;26(6):675-81 - PubMed
  6. Annu Int Conf IEEE Eng Med Biol Soc. 2016 Aug;2016:3402-3405 - PubMed
  7. Clin Neurophysiol. 2007 Dec;118(12):2781-97 - PubMed
  8. Clin Neurophysiol. 2016 May;127(5):2246-56 - PubMed
  9. PLoS One. 2016 Jan 22;11(1):e0145669 - PubMed
  10. J Pediatr. 2009 Sep;155(3):318-23 - PubMed
  11. J Clin Neurophysiol. 2006 Dec;23(6):521-31 - PubMed
  12. Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:4454-7 - PubMed
  13. IEEE Trans Image Process. 2005 Jul;14(7):890-903 - PubMed
  14. J Clin Neurophysiol. 2019 Jan;36(1):9-13 - PubMed
  15. Physiol Meas. 2008 Oct;29(10):1157-78 - PubMed
  16. Int J Neural Syst. 2019 May;29(4):1850030 - PubMed
  17. Clin Neurophysiol. 2006 Jul;117(7):1585-94 - PubMed
  18. Neural Netw. 2020 Mar;123:12-25 - PubMed
  19. Brain. 2014 May;137(Pt 5):1429-38 - PubMed
  20. Clin Neurophysiol. 2007 Jun;118(6):1348-59 - PubMed
  21. Int J Neural Syst. 2019 May;29(4):1850011 - PubMed
  22. Epilepsia. 2006 Jun;47(6):966-80 - PubMed
  23. Epilepsia. 2016 May;57(5):786-95 - PubMed
  24. Med Biol Eng Comput. 2010 Sep;48(9):923-31 - PubMed
  25. Int J Neural Syst. 2016 Dec;26(8):1650027 - PubMed
  26. IEEE Trans Biomed Eng. 2006 Apr;53(4):633-41 - PubMed
  27. Epilepsia. 2005 Jun;46(6):901-17 - PubMed
  28. IEEE Trans Biomed Eng. 2012 Dec;59(12):3379-85 - PubMed
  29. Epilepsy Behav. 2014 Aug;37:291-307 - PubMed
  30. Annu Int Conf IEEE Eng Med Biol Soc. 2017 Jul;2017:2810-2813 - PubMed
  31. Clin Neurophysiol. 2011 Mar;122(3):464-473 - PubMed

Publication Types

Grant support