Identification of Epileptogenic and Non-epileptogenic High-Frequency Oscillations Using a Multi-Feature Convolutional Neural Network Model

Abstract

Accurately identifying epileptogenic zone (EZ) using high-frequency oscillations (HFOs) is a challenge that must be mastered to transfer HFOs into clinical use. We analyzed the ability of a convolutional neural network (CNN) model to distinguish EZ and non-EZ HFOs. Nineteen medically intractable epilepsy patients with good surgical outcomes 2 years after surgery were studied. Five-minute interictal intracranial electroencephalogram epochs of slow-wave sleep were selected randomly. Then 5 s segments of ripples (80–200 Hz) and fast ripples (FRs, 200–500 Hz) were detected automatically. The EZs and non-EZs were identified using the surgery resection range. We innovatively converted all epochs into four types of images using two scales: original waveforms, filtered waveforms, wavelet spectrum images, and smoothed pseudo Wigner–Ville distribution (SPWVD) spectrum images. Two scales were fixed and fitted scales. We then used a CNN model to classify the HFOs into EZ and non-EZ categories. As a result, 7,000 epochs of ripples and 2,000 epochs of FRs were randomly selected from the EZ and non-EZ data for analysis. Our CNN model can distinguish EZ and non-EZ HFOs successfully. Except for original ripple waveforms, the results from CNN models that are trained using fixed-scale images are significantly better than those from models trained using fitted-scale images (p < 0.05). Of the four fixed-scale transformations, the CNN based on the adjusted SPWVD (ASPWVD) produced the best accuracies (80.89 ± 1.43% and 77.85 ± 1.61% for ripples and FRs, respectively, p < 0.05). The CNN using ASPWVD transformation images is an effective deep learning method that can be used to classify EZ and non-EZ HFOs.

[Retracted] Automatic Detection of High‐Frequency Oscillations Based on an End‐to‐End Bi‐Branch Neural Network and Clinical Cross‐Validation

“…Specifically, real HFOs can be identified from the candidate events isolated from background activities by a stacked CNN [ 28 ] or a stacked denoising autoencoder [ 29 ], while in terms of time-frequency diagram, the two-stage automatic detection paradigm has always been used. In particular, it is the simple 2d-CNN structure that is often used as a time-frequency image feature extractor after an initial detector [ 25 , 30 – 32 ]. However, it is still a challenge to fully learn useful information worth attention in either signals or time-frequency images.…”