Discriminating between best performing features for seizure detection and data selection

Abstract

Abstract-Seizure detection algorithms have been developed to solve specific problems, such as seizure onset detection, occurrence detection, termination detection and data selection. It is thus inherent that each type of seizure detection algorithm would detect a different EEG characteristic (feature). However most feature comparison studies do not specify the seizure detection problem for which their respective features have been evaluated. This paper shows that the best features/algorithm bases are not the same for all types of algorithms but depend on the type of seizure detection algorithm wanted. To demonstrate this, 65 features previously evaluated for online seizure data selection are re-evaluated here for seizure occurrence detection, using performance metrics pertinent to each seizure detection type whilst keeping the testing methodology the same. The results show that the best performing features/algorithm bases for data selection and occurrence detection algorithms are different and that it is more challenging to achieve high detection accuracy for the former seizure detection type. This paper also provides a comprehensive evaluation of the performance of 65 features for seizure occurrence detection to aid future researchers in choosing the best performing feature(s) to improve seizure detection accuracy.

PEACOC - Detecting and classifying a wide range of epileptiform activity patterns in rodents

“…Aside from temporal spike statistics (rate, regularity, etc. ), features reflecting amplitudes and signal energy (e. g. line-length), might be interesting since they are popular in seizure detection (Logesparan et al, 2013). Criteria (2) and (3), however, cannot be satisfied easily when using amplitudes.…”

“…Existing spike detection algorithms therefore commonly focus on solitary spikes with high spike amplitudes or clearly discernible waveforms (Anjum et al, 2018; Chauvière et al, 2012; Huneau et al, 2013; Karoly et al, 2016; Krook-Magnuson et al, 2013). Seizure detectors, on the other hand, are mostly designed to specifically detect large, generalized seizures (Esteller et al, 2001; Gardner et al, 2006; Krook-Magnuson et al, 2013; Logesparan et al, 2013). These seizure detectors are often based on measures of signal energy, which can vary substantially depending on signal quality.…”

Improved Patient-Independent System for Detection of Electrical Onset of Seizures

“…They found that the spectral power in low frequencies (delta, theta, and alpha bands) constituted the most reliable signature of the state of consciousness, distinguishing between vegetative state (VS), minimally conscious state, and conscious state (CS). The spectral entropy in the 1 to 45 Hz frequency band showed higher values in minimally conscious and CS than in VS. Logesparan et al 18 analyzed the performance of 65 measures including 17 time domain, 12 Fourier transform, 4 continuous wavelet transform, and 32 discrete wavelet transform based measures, for online data selection and seizure occurrence detection. They concluded that discrete wavelet transform–based power in the 0 to 3.125, 3.125 to 6.25, 6.25 to 12.5, and 12.5 to 25 Hz frequency bands yielded a 0.97 value of the area under the curve in a receiver–operating characteristic analysis and were selected as the best discriminating features for seizure occurrence detection.…”

“…11,12 The continuous discharges of polymorphic waveforms of different amplitude and frequency can be captured using spectral and wavelet features. 13–18 Apart from these, statistical measures 17–19 and entropy measures 20–23 have been proposed to characterize EEG time series. The brain symmetry index of spatial and temporal measures has been proposed to localize the hemisphere of seizure onset, 9,16 and an algorithm using autoregressive spectra has been designed for the detection of temporal lobe seizures.…”

Towards Automatic Identification of Epileptic Recordings in Long-term EEG Monitoring

“…Manual analysis of this data for the identification of epileptic activities is extremely time consuming as the predominant state is the interictal (between seizure) period. Algorithms for detecting seizures have been developed, to reduce the analysis time and aid epileptologists by preselecting only the regions of interest in the EEG signals for review [4], [5], [6]. Such algorithms are known to have accuracy issues in patients with sporadic seizures where they are likely to miss detection of seizures or end up with large number of false positives, and can consequently lead to misdiagnosis.…”