Spike Detection in Biomedical Signal Like Eeg And Ecg Using TEO

Abstract: We propose a novel approach aimed at adaptively setting the threshold of the smoothed Teagor energy operator (STEO) detector to be used in extracellular recording of neural signals. Many types of spike detectors have been proposed all with their advantages and drawbacks. Most of the times there is a trade off between simplicity and performance. The performance of such systems is generally gauged by correct detections and false alarms. The Teagor energy operator is a time frequency analyser that gives high output when both instantaneous amplitude and frequency are high (typical characteristics of the spikes), that is why it is very effective in detecting spikes. The basic TEO gives output after processing three consecutive samples of data, but for signals with high frequency noise it gives more false alarms than the correct detections. To overcome this problem the MTEO (multiresolution TEO) has been proposed. It is observed that for the best performance of any algorithm optimal decision threshold is required. In this we fixed the threshold of MTEO detector based on these parameters and a constant that depends on the tolerance for false alarms. Setting the decision threshold in this way increases the detection performance. If we have some prior knowledge about the spike shape, we can first apply wavelet transform on the signal using suitable mother wavelet and then apply TEO to get better results.

Keywords: Adaptive threshold, robust theory, spike detection, Teagor energy operator.