Latent Phase Detection of Hypoxic-Ischemic Spike Transients in the EEG of Preterm Fetal Sheep Using Reverse Biorthogonal Wavelets & Fuzzy Classifier

Abstract

Hypoxic-ischemic (HI) studies in preterms lack reliable prognostic biomarkers for diagnostic tests of HI encephalopathy (HIE). Our group’s observations from in utero fetal sheep models suggest that potential biomarkers of HIE in the form of developing HI micro-scale epileptiform transients emerge along suppressed EEG/ECoG background during a latent phase of 6–7[Formula: see text]h post-insult. However, having to observe for the whole of the latent phase disqualifies any chance of clinical intervention. A precise automatic identification of these transients can help for a well-timed diagnosis of the HIE and to stop the spread of the injury before it becomes irreversible. This paper reports fusion of Reverse-Biorthogonal Wavelets with Type-1 Fuzzy classifiers, for the accurate real-time automatic identification and quantification of high-frequency HI spike transients in the latent phase, tested over seven in utero preterm sheep. Considerable high performance of 99.78 ± 0.10% was obtained from the Rbio-Wavelet Type-1 Fuzzy classifier for automatic identification of HI spikes tested over 42[Formula: see text]h of high-resolution recordings (sampling-freq:1024[Formula: see text]Hz). Data from post-insult automatic time-localization of high-frequency HI spikes reveals a promising trend in the average rate of the HI spikes, even in the animals with shorter occlusion periods, which highlights considerable higher number of transients within the first 2[Formula: see text]h post-insult.

Latent Phase Identification of High-Frequency Micro-Scale Gamma Spike Transients in the Hypoxic Ischemic EEG of Preterm Fetal Sheep Using Spectral Analysis and Fuzzy Classifiers

“…The negligible accuracy drop of <1% is within an acceptable marginal error-tolerance range and can be accepted when the algorithm is using a simpler technique. Moreover, the current article provides supplementary spectral analysis materials based on the wavelet power spectral density and pseudo-frequency approximation analysis techniques that mathematically justify the choice of frequency band of 80-120 Hz for HI spike transients and further confirm the correct choice of the mother wavelet and the scale number in our previous work [20]. The complementary spectral material in this article will also further explain the very close performance similarities between the FFT-based approach (in this work) compared to the wavelet-based strategy in the previous work.…”

“…[5][6][7][8][14][15][16][17][18]. The micro-scale EEG transients are mainly categorized as spikes, sharps, slow waves, and complexes, as well as ongoing stereotypic evolving micro-scale seizures (SEMS) [5][6][7][8][9][10][11][14][15][16]19] (see Figure 2 of ref [20]). Early automatic identification of these transients could support in-time diagnosis of at-risk infants and increase the chances of effective treatment before the window of opportunity is closed.…”

“…We have recently developed automated algorithms for the accurate identification of HI spikes and sharp waves [16][17][18]20,21] and demonstrated that an increase in the number of automatically identified micro-scale sharp waves from 2-6 h after HI is associated with worse fetal outcomes [9,10,21]. Monitoring of post-asphyxial EEG from instrumented preterm fetal sheep demonstrates that micro-scale HI spike transients emerge in the gamma frequency band (namely 80-120 Hz) along the 1024 Hz high-frequency sampled electrocorticogram (ECoG) within the latent phase (Figure 1).…”

“…Using 1024 Hz sampled ECoGs from seven in utero preterm fetal sheep, we recently demonstrated that the fusion of thresholded CWT (using the reverse biorthogonal wavelets rbio2.8 basis function) and type-1 fuzzy classifier (rbio-WT-Type-1-FLC) significantly improves the accuracy of the spike identification in real time [20]. Using data from a similar database, this paper represents a new fusion technique based on spectral Fourier analysis and Type-I fuzzy classifiers (FFT-Type-I FLC classifier) for the accurate real-time automatic identification and quantification of high-frequency micro-scale HI spike transients (in the gamma frequency band, namely 80-120 Hz, in the latent phase) in high-resolution ECoG recordings, post insult.…”

“…Using data from a similar database, this paper represents a new fusion technique based on spectral Fourier analysis and Type-I fuzzy classifiers (FFT-Type-I FLC classifier) for the accurate real-time automatic identification and quantification of high-frequency micro-scale HI spike transients (in the gamma frequency band, namely 80-120 Hz, in the latent phase) in high-resolution ECoG recordings, post insult. The article describes the possibility of replacing the CWT block of the previous rbio-WT-Type-1-FLC classifier in [20] with a more straightforward spectral Fourier analysis block that leads to competitive high accuracies compared to the previous wavelet-based technique. The negligible accuracy drop of <1% is within an acceptable marginal error-tolerance range and can be accepted when the algorithm is using a simpler technique.…”