TY  - CONF
  - Faul S.; Boylan G. B.; Connolly S.; Marnane W. P.; Lightbody G.
  - The Third IASTED International Conference on Signal Processing, Pattern Recognition and Applications (SPPRA 2006)
  - Gaussian Process Modelling as an Indicator of Neonatal Seizures
  - 2006
  - February
  - Published
  - 1
  - ()
  - 177
  - 182
  - Insbruck, Austria
  - 15-FEB-06
  - 17-FEB-06
  - Gaussian process models have some attractive advantages over parametric models and neural networks. They have a small number of tunable parameters, give a measure of the uncertainty of the model prediction, and obtain a rel atively good model when only a small set of training data is available. In this study the theory of Gaussian process models has been applied to the neonatal seizure detection problem. Two measures are calculated from 1 second win dows of EEG recordings; the variance (certainty) of a one step ahead prediction, and the ratio of the ;#64257;rst model hy perparameter to the last. In ANOVA tests both measures show statistical difference in their values for non-seizure and seizure EEG. A comparison with a similar Autoregres sive (AR) modelling approach shows that Gaussian Process model methods show great promise in real-time neonatal seizure detection.
  - http://www.iasted.org/conferences/pastinfo-520.html
DA  - 2006/02
ER  - 

@inproceedings{V377368,
   = {Faul S. and  Boylan G. B. and  Connolly S. and  Marnane W. P. and  Lightbody G.},
   = {The Third IASTED International Conference on Signal Processing, Pattern Recognition and Applications (SPPRA 2006)},
   = {{Gaussian Process Modelling as an Indicator of Neonatal Seizures}},
   = {2006},
   = {February},
   = {Published},
   = {1},
   = {()},
  pages = {177--182},
   = {Insbruck, Austria},
  month = {Feb},
   = {17-FEB-06},
   = {{Gaussian process models have some attractive advantages over parametric models and neural networks. They have a small number of tunable parameters, give a measure of the uncertainty of the model prediction, and obtain a rel atively good model when only a small set of training data is available. In this study the theory of Gaussian process models has been applied to the neonatal seizure detection problem. Two measures are calculated from 1 second win dows of EEG recordings; the variance (certainty) of a one step ahead prediction, and the ratio of the ;#64257;rst model hy perparameter to the last. In ANOVA tests both measures show statistical difference in their values for non-seizure and seizure EEG. A comparison with a similar Autoregres sive (AR) modelling approach shows that Gaussian Process model methods show great promise in real-time neonatal seizure detection.}},
   = {http://www.iasted.org/conferences/pastinfo-520.html},
  source = {IRIS}
}