Channel Increment Strategy-Based 1D Convolutional Neural Networks for Seizure Prediction Using Intracranial EEG

Abstract

The application of intracranial electroencephalogram (iEEG) to predict seizures remains challenging. Although channel selection has been utilized in seizure prediction and detection studies, most of them focus on the combination with conventional machine learning methods. Thus, channel selection combined with deep learning methods can be further analyzed in the field of seizure prediction. Given this, in this work, a novel iEEG-based deep learning method of One-Dimensional Convolutional Neural Networks (1D-CNN) combined with channel increment strategy was proposed for the effective seizure prediction. First, we used 4-sec sliding windows without overlap to segment iEEG signals. Then, 4-sec iEEG segments with an increasing number of channels (channel increment strategy, from one channel to all channels) were sequentially fed into the constructed 1D-CNN model. Next, the patient-specific model was trained for classification. Finally, according to the classification results in different channel cases, the channel case with the best classification rate was selected for each patient. Our method was tested on the Freiburg iEEG database, and the system performances were evaluated at two levels (segment-and event-based levels). Two model training strategies (Strategy-1 and Strategy-2) based on the K-fold cross validation (K-CV) were discussed in our work. (1) For the Strategy-1, a basic K-CV, a sensitivity of 90.18%, specificity of 94.81%, and accuracy of 94.42% were achieved at the segment-based level. At the event-based level, an eventbased sensitivity of 100%, and false prediction rate (FPR)

Efficient Seizure Prediction and EEG Channel Selection Based on Multi-Objective Optimization

“…Hence, 2D-CNN and 3D-CNN will not be suitable to implement low-power/low-memory devices. The 1D-CNN can achieve excellent performance in several applications [21], [40]. Our main aim is to implement a seizure prediction device which will be low-cost hardware.…”

“…Hence, most of the researchers used deep learning techniques, which were used to extract the most relevant features from the EEG signals for the classification of pre-ictal and inter-ictal states to predict seizures in advance. The most commonly used deep learning techniques are Convolution neural networks [19], [20], [21], LSTM [22], [23], [24], DenseNet [24], [25], Self-Organizing Maps [26], and Long-term recurrent convolutional networks [27]. It was found that the use of deep learning techniques provided better accuracy for seizure prediction.…”