Conference Publication Details
Mandatory Fields
Ahmed R.;Temko A.;Marnane W.;Boylan G.;Lightbody G.
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Grading brain injury in neonatal EEG using SVM and supervector kernel
2014
January
Validated
1
Scopus: 4 ()
Optional Fields
5894
5898
Brain injury at the time of birth could lead to severe neurological dysfunction at an older age. Grading the brain injury in the early hours after birth could help doctors determine a prompt and reliable treatment. This work presents an automated neonatal EEG grading system based on a cross-disciplinary method of using Support Vector Machine and supervectors, initially developed for speaker identification. The EEG is classified into one of the four grades of neonatal brain injury. The preliminary results show promising performance and are an improvement on the previously published results. © 2014 IEEE.
10.1109/ICASSP.2014.6854734
Grant Details