Automatic Localization of Epileptogenic Zone Using Nonlinear Envelope Differential Instantaneous Energy Detector and High-Frequency Oscillation Analysis

This study introduces a frequency-weighted nonlinear Envelope derivative energy detector combined with a Staba method for spotting high-frequency oscillations (HFOs) by multichannel intracranial electroencephalogram (iEEG) signals automatically and hence to perform seizure zone localization by calculating HFO rate/min. First, the energy of the iEEG signal is detected by a non-negative instantaneous energy operator and applied as an input to the staba method for detecting HFOs, distinguishing HFOs from irrelevant activities. It is essential to include information based on amplitude and frequency to analyze iEEG signals more efficiently.In iEEG analysis, the Teager detector is employed in various energy-based biomedical applications using the advantage of being frequency-weighted. In signal processing aspects, the Teager-kaiser operator is meagerly described due to the negative modulation term present in the detector. We propose an instantaneous energy detector that is non-negative and uses the derivative of an envelope of the signal. In addition, this operator is less sensitive to noise and quantization errors than the Teiger-kaiser and Agarwal Gottman operators. Moreover, we don’t need to post-process the output of the envelope derivative detector to extricate the frequency and amplitude information. A comparison unveils that our energy locater combined with the Staba method came up with superior localization attainment in accuracy, sensitivity, and specificity than Teager-Kaiser and Agarwal-Gotman operators.