Lightweight convolution transformer for cross-patient seizure detection in multi-channel EEG signals

Rukhsar, Salim; Tiwari, Anil Kumar

doi:10.1016/j.cmpb.2023.107856

Cited by 7 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Seizure detection using EEG signals has witnessed remarkable advancements with the application of deep learning techniques [8,9,10,11]. In recent years, numerous studies have explored the utilization of sophisticated neural network architectures such as LSTM networks [12,13,6] networks and Transformers [5,14] for enhancing the accuracy and reliability of seizure detection and prediction. This section presents an overview of recent publications that exemplify the contributions made in this direction.…”

Section: Related Workmentioning

confidence: 99%

“…Recent advancements in the application of Transformers to sequential data have extended to EEG analysis for seizure detection and predictions. Recently, a lightweight Transformer architecture has been proposed to enhance accuracy in detecting seizure patterns by capturing local and global dependencies within EEG signals [5]. Similarly, [14] introduced a Multichannel Vision Transformer (MViT) for automated spatiotemporal spectral features learning in multi-channel EEG data to achieve state-of-the-art performance in seizure prediction.…”

Section: Related Workmentioning

confidence: 99%

“…A recent study [19] introduced a hybrid model that employs a 1D Convolutional Neural Network (CNN) with a Bidirectional Long Short-Term Memory (BiLSTM) network to detect seizures. Convolutional tokenizer is used combined with ViT to effectively capture the temporal and spatial information in time series dataset [5,20]. In another work [21], R-Transformer is proposed where they used LocalRNN for local dependencies in series with a selfattention mechanism to capture global dependencies.…”

Section: Related Workmentioning

confidence: 99%

“…Despite the stated benefits, Transformers do have drawbacks. Processing lengthy documents, such as technical articles or ultra-high pixel images, is hampered by the quadratic computational complexity of self-attention with respect to the sequence length [2,5,6]. Original Transformer architecture inherently does not understand or handle the relationships and interactions between different channels or variables in multichannel time series data.…”

Section: Introductionmentioning

confidence: 99%

“…This ARNN cell architecture is capable of handling multi- channel EEG signals, that have dominantly high frequency bursts with a long-term relation in seizure classification [7,5]. By combining local and global features, the model gains access to a rich representation of the input data, which leads to better overall detection performance.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Deep Optimized Electrodes and Frequency Bands in the Phase Space for Identification of Seizures

Rukhsar

Tiwari

Panda

2022

2022 IEEE 19th India Council International Conference (INDICON)

View full text Add to dashboard Cite

The electroencephalogram (EEG) has been extensively used to record brain signals for the analysis of neurological diseases over several decades. Epilepsy is the one of the most common neurological disorder in the world. It is associated with seizures, which are a clinical manifestation of abnormal, excessive excitation and synchronisation of a population of cortical neurons, resulting in abrupt changes in consciousness, movement, behaviour, and feelings. Our goal in this letter is to propose a computationally efficient EEG based system that can detect seizures with improved detection performance. To deeply optimize the electrodes and frequency bands, we proposed a fast computational algorithm by finding the critical electrodes and optimal sub-bands of EEG signals. We used wavelet packet decomposition (WPD) to find optimal sub-bands and maximum relevance minimum redundancy (mRMR) method for electrodes selection. It is found that the portrait of the features in the phase space depicted promising discriminating ability to detect seizure efficiently. Experimental results show that the proposed method outperformed the state-of-art methods and provided 100% and 94.79% of detection accuracy on UBonn and CHB-MIT datasets, respectively, at a significantly reduced computational cost.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Deep Optimized Electrodes and Frequency Bands in the Phase Space for Identification of Seizures

Rukhsar

Tiwari

Panda

2022

2022 IEEE 19th India Council International Conference (INDICON)

View full text Add to dashboard Cite

show abstract

ARNN: Attentive recurrent neural network for multi-channel EEG signals to identify epileptic seizures

Rukhsar,

Tiwari

2025

Neurocomputing

View full text Add to dashboard Cite

Signal to Image Conversion and Convolutional Neural Networks for Physiological Signal Processing: A Review

Ch Vidyasagar,

Revanth Kumar,

Anantha Sai

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Physiological signals such as electroencephalography (EEG), electromyography (EMG), and electrocardiography (ECG) provide valuable clinical information but pose challenges for analysis due to their high-dimensional nature. Traditional machine learning techniques, relying on hand-crafted features from fixed analysis windows, can lead to the loss of discriminative information. Recent studies have demonstrated the effectiveness of deep convolutional neural networks (CNNs) for robust automated feature learning from raw physiological signals. However, standard CNN architectures require two-dimensional image data as input. This has motivated research into innovative signal-to-image (STI) transformation techniques to convert one-dimensional time series into images preserving spectral, spatial, and temporal characteristics. This paper reviews recent advances in strategies for physiological signal-to-image conversion and their applications using CNNs for automated processing tasks. A systematic analysis of EEG, EMG, and ECG signal transformation and CNN-based analysis techniques spanning diverse applications, including brain-computer interfaces, seizure detection, motor control, sleep stage classification, arrhythmia detection, and more, are presented. Key insights are synthesised regarding the relative merits of different transformation approaches, CNN model architectures, training procedures, and benchmark performance. Current challenges and promising research directions at the intersection of deep learning and physiological signal processing are discussed. This review aims to catalyse continued innovations in effective end-to-end systems for clinically relevant information extraction from multidimensional physiological data using deep neural networks by providing a comprehensive overview of state-of-the-art techniques.

show abstract

Lightweight convolution transformer for cross-patient seizure detection in multi-channel EEG signals

Cited by 7 publications

References 37 publications

Deep Optimized Electrodes and Frequency Bands in the Phase Space for Identification of Seizures

Deep Optimized Electrodes and Frequency Bands in the Phase Space for Identification of Seizures

ARNN: Attentive recurrent neural network for multi-channel EEG signals to identify epileptic seizures

Signal to Image Conversion and Convolutional Neural Networks for Physiological Signal Processing: A Review

Contact Info

Product

Resources

About