Robust Multichannel EEG Compressed Sensing in the Presence of Mixed Noise

Li, Chang; Tao, Weiliang; Cheng, Juan; Liu, Yu; Chen, Xun

doi:10.1109/jsen.2019.2930546

Cited by 23 publications

(4 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The singlechannel method achieved up to a 4% improvement in the area under the curve, and the multivariate feature achieved high mean AUC values of 94.1%. Li et al [60] introduced the Sparse and Low-Rank Representation in the presence of Mixed Noise (SLRMN) method for robust multichannel EEG signal compression. The SLRMN method improved the accuracy of compressed signal recovery in the presence of mixed noise.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Compressive Sensing and Classification Approach for Dynamic Storage Management of Vital Biomedical Signals

Emara,

El-Shafai,

Algarni

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The efficient compression and classification of medical signals, particularly electroencephalography (EEG) and electrocardiography (ECG) signals in wireless body area network (WBAN) systems, are crucial for real-time monitoring and diagnosis. This paper addresses the challenges of compressive sensing and classification in WBAN systems for EEG and ECG signals. To tackle these challenges, a sequential approach is proposed. The first step involves compressing the EEG and ECG signals using the optimized Walsh-Hadamard transform (OWHT). This transform allows for efficient representation of the signals while preserving their essential characteristics. However, the presence of noise can impact the quality of the compressed signals. To mitigate this effect, the signals are subsequently recovered using the Sparse Group Lasso 1 (SPGL1) algorithm and OWHT, which take into account the noise characteristics during the recovery process. To evaluate the performance of the proposed compressive sensing algorithm, two metrics are employed: mean squared error (MSE) and maximum correntropy criterion (MCC). These metrics provide insights into the accuracy and reliability of the recovered signals at different signal-to-sample ratio (SSR) ratios. The results of the evaluation demonstrate the effectiveness of the proposed algorithm in accurately reconstructing the EEG and ECG signals while effectively managing the noise interference. Furthermore, to enhance the classification capabilities of the compressed signals, a local binary pattern (LBP) approach is applied. This technique extracts discriminative features from the compressed signals, which are then fed into a classification algorithm based on residual learning. This classification algorithm is trained from scratch and specifically designed to work with the compressed signals. The experimental results showcase the high accuracy achieved by the proposed algorithm in classifying the compressed EEG and ECG signals without the need for signal recovery. The findings of this study highlight the potential of the proposed approach in achieving efficient and accurate medical signal analysis in WBAN systems. By eliminating the computational burden of signal recovery and leveraging the advantages of compressive sensing, the proposed algorithm offers a promising solution for real-time monitoring and diagnosis, ultimately improving the overall efficiency and effectiveness of healthcare systems.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hybrid Compressive Sensing and Classification Approach for Dynamic Storage Management of Vital Biomedical Signals

Emara,

El-Shafai,

Algarni

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This can be attributed to its non-invasive nature, prompt response, and high temporal resolution [7]. Consequently, EEG-based emotion recognition has become a central focus in the field of emotion recognition research [8].…”

Section: Introductionmentioning

confidence: 99%

AutoEER: automatic EEG-based emotion recognition with neural architecture search

Liu

Zhang

et al. 2023

J. Neural Eng.

View full text Add to dashboard Cite

Objective. Emotion recognition based on electroencephalography (EEG) is garnering increasing attention among researchers due to its wideranging applications and the rise of portable devices. Deep learning-based models have demonstrated impressive progress in EEG-based emotion recognition, thanks to their exceptional feature extraction capabilities. However, the manual design of deep networks is time-consuming and labour-intensive. Moreover, the inherent variability of EEG signals necessitates extensive customization of models, exacerbating these challenges. Neural architecture search (NAS) methods can alleviate the need for excessive manual involvement by automatically discovering the optimal network structure for EEG-based emotion recognition.Approach. In this regard, we propose AutoEER (Automatic EEG-based Emotion Recognition), a framework that leverages tailored NAS to automatically discover the optimal network structure for EEG-based emotion recognition. We carefully design a customized search space specifically for EEG signals, incorporating operators that effectively capture both temporal and spatial properties of EEG.Additionally, we employ a novel parameterization strategy to derive the optimal network structure from the proposed search space. Main results. Extensive experimentation on emotion classification tasks using two benchmark datasets, DEAP and SEED, has demonstrated that AutoEER outperforms state-of-the-art (SOTA) manual deep and NAS models. Specifically, compared to the optimal model WangNAS on the accuracy (ACC) metric, AutoEER improves its average accuracy on all datasets by 0.93%. Similarly, compared to the optimal model LiNAS on the F1 Ssore (F1) metric, AutoEER improves its average F1 score on all datasets by 4.51%. Furthermore, the architectures generated by AutoEER exhibit superior transferability compared to alternative methods. Significance.AutoEER represents a novel approach to EEG analysis, utilizing a specialized search space to design models tailored to individual subjects. This approach significantly reduces the labour and time costs associated with manual model construction in EEG research, holding great promise for advancing the field and streamlining research practices.

show abstract

“…Therefore, accurate and reliable recognition of EEG signals [ 11 ] contributes to improving the precision and credibility of systems. In recent years, researchers have extensively studied emotion recognition based on EEG signals using machine learning and deep learning methods [ 12 , 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

EEG Emotion Recognition by Fusion of Multi-Scale Features

Du,

Meng,

Qiu

et al. 2023

Brain Sciences

View full text Add to dashboard Cite

Electroencephalogram (EEG) signals exhibit low amplitude, complex background noise, randomness, and significant inter-individual differences, which pose challenges in extracting sufficient features and can lead to information loss during the mapping process from low-dimensional feature matrices to high-dimensional ones in emotion recognition algorithms. In this paper, we propose a Multi-scale Deformable Convolutional Interacting Attention Network based on Residual Network (MDCNAResnet) for EEG-based emotion recognition. Firstly, we extract differential entropy features from different channels of EEG signals and construct a three-dimensional feature matrix based on the relative positions of electrode channels. Secondly, we utilize deformable convolution (DCN) to extract high-level abstract features by replacing standard convolution with deformable convolution, enhancing the modeling capability of the convolutional neural network for irregular targets. Then, we develop the Bottom-Up Feature Pyramid Network (BU-FPN) to extract multi-scale data features, enabling complementary information from different levels in the neural network, while optimizing the feature extraction process using Efficient Channel Attention (ECANet). Finally, we combine the MDCNAResnet with a Bidirectional Gated Recurrent Unit (BiGRU) to further capture the contextual semantic information of EEG signals. Experimental results on the DEAP dataset demonstrate the effectiveness of our approach, achieving accuracies of 98.63% and 98.89% for Valence and Arousal dimensions, respectively.

show abstract

Robust Multichannel EEG Compressed Sensing in the Presence of Mixed Noise

Cited by 23 publications

References 54 publications

A Hybrid Compressive Sensing and Classification Approach for Dynamic Storage Management of Vital Biomedical Signals

A Hybrid Compressive Sensing and Classification Approach for Dynamic Storage Management of Vital Biomedical Signals

AutoEER: automatic EEG-based emotion recognition with neural architecture search

EEG Emotion Recognition by Fusion of Multi-Scale Features

Contact Info

Product

Resources

About