Parallel Ictal-Net, a Parallel CNN Architecture with Efficient Channel Attention for Seizure Detection

Gerardo Hernández-Nava,Sebastián Salazar-Colores,Eduardo Cabal-Yepez,Juan-Manuel Ramos-Arreguín

DOI: https://doi.org/10.3390/s24030716

IF: 3.9

2024-01-23

Sensors

Abstract:Around 70 million people worldwide are affected by epilepsy, a neurological disorder characterized by non-induced seizures that occur at irregular and unpredictable intervals. During an epileptic seizure, transient symptoms emerge as a result of extreme abnormal neural activity. Epilepsy imposes limitations on individuals and has a significant impact on the lives of their families. Therefore, the development of reliable diagnostic tools for the early detection of this condition is considered beneficial to alleviate the social and emotional distress experienced by patients. While the Bonn University dataset contains five collections of EEG data, not many studies specifically focus on subsets D and E. These subsets correspond to EEG recordings from the epileptogenic zone during ictal and interictal events. In this work, the parallel ictal-net (PIN) neural network architecture is introduced, which utilizes scalograms obtained through a continuous wavelet transform to achieve the high-accuracy classification of EEG signals into ictal or interictal states. The results obtained demonstrate the effectiveness of the proposed PIN model in distinguishing between ictal and interictal events with a high degree of confidence. This is validated by the computing accuracy, precision, recall, and F1 scores, all of which consistently achieve around 99% confidence, surpassing previous approaches in the related literature.

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation

What problem does this paper attempt to address?

The main objective of this paper is to develop a method that can efficiently distinguish between ictal and interictal states in EEG signals of epilepsy patients. To achieve this goal, the research team proposed the Parallel Ictal-Net (PIN), a parallel convolutional neural network (CNN) architecture that uses spectrograms obtained from continuous wavelet transform (CWT) as input and incorporates an efficient channel attention mechanism. Specifically, the paper aims to address the following issues: 1. **Improve epilepsy detection accuracy**: Given the impact of epilepsy on patients and their families, it is crucial to develop reliable and efficient auxiliary tools for timely diagnosis. Therefore, this study aims to improve the accuracy of epilepsy detection by proposing a new method. 2. **Distinguish between ictal and interictal states**: Using subsets D (ictal events) and E (interictal events) from the publicly available Bonn University dataset, which contain data collected from the epileptic region, the focus of the research is to accurately distinguish between these two states. 3. **Enhance the performance of existing technologies**: Although existing studies have achieved good results, there is still room for improvement. The PIN architecture aims to surpass previous methods, achieving approximately 99% confidence levels in classification accuracy, precision, recall, and F1 score. In summary, this paper is dedicated to addressing one of the key challenges in the field of epilepsy detection—accurately identifying ictal and interictal states through the analysis of EEG signals to improve patient treatment plans and quality of life.

Parallel Ictal-Net, a Parallel CNN Architecture with Efficient Channel Attention for Seizure Detection

Real-Time Epileptic Seizure Detection Based on Deep Learning.

Epilepsy Seizure Prediction on EEG Using Common Spatial Pattern and Convolutional Neural Network.

Patient-Independent Seizure Detection Based on Channel-Perturbation Convolutional Neural Network and Bidirectional Long Short-Term Memory

Epileptic-Net: An Improved Epileptic Seizure Detection System Using Dense Convolutional Block with Attention Network from EEG

Evaluating CNN Methods for Epileptic Seizure Type Classification Using EEG Data

Convolutional Neural Network for Seizure Detection of Nocturnal Frontal Lobe Epilepsy

SeizureNet: Multi-Spectral Deep Feature Learning for Seizure Type Classification

Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals

Epileptic Seizure Prediction Using Attention Augmented Convolutional Network

Parallel Dual-Branch Fusion Network for Epileptic Seizure Prediction

CNN-Informer: A hybrid deep learning model for seizure detection on long-term EEG

Patient Independent Interictal Epileptiform Discharge Detection

Permutation Entropy-Based Interpretability of Convolutional Neural Network Models for Interictal EEG Discrimination of Subjects with Epileptic Seizures vs. Psychogenic Non-Epileptic Seizures

Basic study of epileptic seizure detection using a single-channel frontal EEG and a pre-trained ResNet

MP-SeizNet: A Multi-Path CNN Bi-LSTM Network for Seizure-Type Classification Using EEG

A Novel Multi-scale Dilated 3D CNN for Epileptic Seizure Prediction

Preictal and Interictal Recognition for Epileptic Seizure Prediction Using Pre-trained 2D-CNN Models

EEG-Based Seizure detection using linear graph convolution network with focal loss

Automatic interictal epileptiform discharge (IED) detection based on convolutional neural network (CNN)

Biological plausible algorithm for seizure detection: Toward AI-enabled electroceuticals at the edge