Exploring the Applicability of Transfer Learning and Feature Engineering in Epilepsy Prediction Using Hybrid Transformer Model

Hu, Shuaicong; Liu, Jian; Yang, Rui Q.; Wang, Ya’nan; Wang, Aiguo; Li, Kuanzheng; Liu, Wenxin; Yang, Cuiwei

doi:10.1109/tnsre.2023.3244045

Cited by 21 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 5 compares our method to methods developed by other authors using only the CHB-MIT dataset. The proposed method outperforms several state-of-the-art methods for detecting epilepsy from EEG signals, specifically for the study in [4,[6][7][8][9][10][31][32][33][34][35][36][37][38][39]. It can be seen that the accuracy of the method that only uses CNN without feature extraction is lower than the accuracy of the method we propose, whereas the accuracy of the method that uses feature extraction such as DWT and FFT is higher than 90%.…”

Section: Comparison With Existing Methodsmentioning

confidence: 81%

Optimized One-Dimension Convolutional Neural Network for Seizure Classification from EEG Signal based on Whale Optimization Algorithm

2023

IJIES

View full text Add to dashboard Cite

Epilepsy is a chronic disorder that causes sudden, recurring seizures and early detection of seizures is needed for prompt treatment to reduce the higher risk. An electroencephalogram (EEG) can detect epilepsy based on traces of electrical activity and wave patterns in the brain. However, analyzing EEG signals takes a long time and is operated by neuroscientists. In this paper, we propose automatic seizure detection using a one-dimension convolutional neural network (1D CNN) and the approach of whale optimization algorithm (WOA). The EEG signal is trimmed every three seconds, and features are extracted using discrete wavelet transform (DWT). The WOA approach was used to optimize the number of layers and neurons in 1D CNN. The experimental results show that the proposed model can improve CNN's performance in detecting seizures with an accuracy of 99.76%, respectively. The proposed method is suitable for the children's hospital bostonmassachusetts institute of technology (CHB-MIT) dataset.

show abstract

Section: Comparison With Existing Methodsmentioning

confidence: 81%

Optimized One-Dimension Convolutional Neural Network for Seizure Classification from EEG Signal based on Whale Optimization Algorithm

2023

IJIES

View full text Add to dashboard Cite

show abstract

“…CNNs are DL models inspired by a human's nervous visual processing. These networks are competent in analyzing high-dimensional data, which may include graphs or images [ 35 ]. CNNs are multi-layered and include convolutional, fully connected, and pooling layers.…”

Section: Reviewmentioning

confidence: 99%

The Application of Deep Learning to Electroencephalograms, Magnetic Resonance Imaging, and Implants for the Detection of Epileptic Seizures: A Narrative Review

Singh¹,

Velagala²,

Kumar³

et al. 2023

Cureus

View full text Add to dashboard Cite

Epilepsy is a neurological disorder characterized by recurrent seizures affecting millions worldwide. Medically intractable seizures in epilepsy patients are not only detrimental to the quality of life but also pose a significant threat to their safety. Outcomes of epilepsy therapy can be improved by early detection and intervention during the interictal window period. Electroencephalography is the primary diagnostic tool for epilepsy, but accurate interpretation of seizure activity is challenging and highly time-consuming. Machine learning (ML) and deep learning (DL) algorithms enable us to analyze complex EEG data, which can not only help us diagnose but also locate epileptogenic zones and predict medical and surgical treatment outcomes. DL models such as convolutional neural networks (CNNs), inspired by visual processing, can be used to classify EEG activity. By applying preprocessing techniques, signal quality can be enhanced by denoising and artifact removal. DL can also be incorporated into the analysis of magnetic resonance imaging (MRI) data, which can help in the localization of epileptogenic zones in the brain. Proper detection of these zones can help in good neurosurgical outcomes. Recent advancements in DL have facilitated the implementation of these systems in neural implants and wearable devices, allowing for real-time seizure detection. This has the potential to transform the management of drug-refractory epilepsy. This review explores the application of ML and DL techniques to Electroencephalograms (EEGs), MRI, and wearable devices for epileptic seizure detection. This review briefly explains the fundamentals of both artificial intelligence (AI) and DL, highlighting these systems' potential advantages and undeniable limitations.

show abstract

“…Increasing scientific advancements in optimization methods, signal processing methods, machine learning and transfer learning, and deep learning in seizure and epilepsy classification have been observed (Chavan & Desai, 2023;Luo et al, 2022;Rasheed et al, 2020;Shoeibi et al, 2021;Singh & Malhotra, 2022c;Supriya et al, 2020;Tang et al, 2023). Newer studies have also focused on automatic epilepsy prediction (Altaf et al, 2023;Shuaicong et al, 2023;Singh & Malhotra, 2021a;Singh & Malhotra, 2021b;Wang et al, 2022;Xin et al, 2023). Recent concepts in meta-learning, short learning and meta-transfer learning have not been explored in this field.…”

Section: Chb-mit Scalp Eeg Database Collected From Children's Hospitalmentioning

confidence: 99%

A review on software and hardware developments in automatic epilepsy diagnosis using EEG datasets

et al. 2023

View full text Add to dashboard Cite

Epilepsy is a common non-communicable, group of neurological disorders affecting more than 50 million individuals worldwide. Different approaches of basic, clinical, and translational research of the human brain have been explored to diagnose, treat, and manage the growing no. of cases of epilepsy. Various hospital information from video, images, signals, forms, and so forth, are retrieved and analysed to develop a consensus for such patients. Electroencephalography (EEG) tests are routinely used to diagnose the type of epilepsy in a clinical setting. Artificial Intelligence algorithms are assisting in the early detection and prediction of epileptic patterns observed in EEG signals. This paper reviews recent and emerging state-of-the-art (SOTA) software and hardware approaches in data selection, signal processing, feature estimation, classification, detection methods, and evaluation metrics applied to open and private EEG datasets from 2014 to 2022. The work summarizes and compares reported works through subjective and objective parameters. Rise in hardware pipeline, start-ups, and companies, deep learning methods in software pipeline and release of free EEG datasets have been observed from 2014 to 2022. SOTA

show abstract

Exploring the Applicability of Transfer Learning and Feature Engineering in Epilepsy Prediction Using Hybrid Transformer Model

Cited by 21 publications

References 39 publications

Optimized One-Dimension Convolutional Neural Network for Seizure Classification from EEG Signal based on Whale Optimization Algorithm

Optimized One-Dimension Convolutional Neural Network for Seizure Classification from EEG Signal based on Whale Optimization Algorithm

The Application of Deep Learning to Electroencephalograms, Magnetic Resonance Imaging, and Implants for the Detection of Epileptic Seizures: A Narrative Review

A review on software and hardware developments in automatic epilepsy diagnosis using EEG datasets

Contact Info

Product

Resources

About