Real-Time Seizure Detection Based on EEG and ECG Fused Features Using Gabor Functions

using the scalp electroencephalogram (EEG) to detect seizure onsets that are not associated with rhythmic EEG activity is challenging. In this paper, we illustrate how supplementing the extracted information from the scalp EEG with the extracted information from electrocardiogram (ECG) can improve the detection of these types of seizures. In this scheme, spectral and spatial features are extracted from EEG and ECG signals by Gabor functions. Then a k-nearest neighbour classifier is used to classify the extracted features from seizure and non-seizure EEG-ECG signals. This algorithm can automatically detect the presence of seizures which can be important advance facilitating timely medical intervention. The performance of algorithm is evaluated on 12 records and recognizes 98.31% expert-labeled seizures with a false detection rate of 11.52%.

[1]  Arun K. Pujari,et al.  A modified Gabor function for content based image retrieval , 2007, Pattern Recognit. Lett..

[2]  Touradj Ebrahimi,et al.  Classification of EEG signals using Dempster Shafer theory and a k-nearest neighbor classifier , 2009, 2009 4th International IEEE/EMBS Conference on Neural Engineering.

[3]  C. DeGiorgio,et al.  Pilot Study of Trigeminal Nerve Stimulation (TNS) for Epilepsy: A Proof‐of‐Concept Trial , 2006, Epilepsia.

[4]  Kenji Okajima,et al.  The Gabor function extracts the maximum information from input local signals , 1998, Neural Networks.

[5]  Reza Tafreshi,et al.  Automated Real-Time Epileptic Seizure Detection in Scalp EEG Recordings Using an Algorithm Based on Wavelet Packet Transform , 2010, IEEE Transactions on Biomedical Engineering.

[6]  Sheng-Fu Liang,et al.  A Portable Wireless Online Closed-Loop Seizure Controller in Freely Moving Rats , 2011, IEEE Transactions on Instrumentation and Measurement.

[7]  E. Rodriguez-Villegas,et al.  Algorithm for AEEG data selection leading to wireless and long term epilepsy monitoring , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[8]  K. Gröchenig.  Multivariate Gabor frames and sampling of entire functions of several variables , 2011 .

[9]  Esther Rodriguez-Villegas,et al.  Wearable Electroencephalography , 2010, IEEE Engineering in Medicine and Biology Magazine.

[10]  P16.1 Neonatal seizure detection using a combination of EEG and ECG analysis , 2006, Clinical Neurophysiology.

[11]  Paul B. Colditz,et al.  Signal processing applications in clinical newborn medicine to improve health outcomes , 2010, 10th International Conference on Information Science, Signal Processing and their Applications (ISSPA 2010).

[12]  Paul B. Colditz,et al.  A computer-aided detection of EEG seizures in infants: a singular-spectrum approach and performance comparison , 2002, IEEE Transactions on Biomedical Engineering.

[13]  H. Khamis,et al.  Seizure state detection of temporal lobe seizures by autoregressive spectral analysis of scalp EEG , 2009, Clinical Neurophysiology.

[14]  R. Reilly,et al.  Combination of EEG and ECG for improved automatic neonatal seizure detection , 2007, Clinical Neurophysiology.

[15]  J. Gotman.  Automatic seizure detection: improvements and evaluation. , 1990, Electroencephalography and clinical neurophysiology.