TSception:A Deep Learning Framework for Emotion Detection Using EEG

Ding, Yi; Robinson, Neethu; Zeng, Qiuhao; Chen, Duo; Wai, Aung Aung Phyo; Lee, Tih-Shih; Guan, Cuntai

doi:10.1109/ijcnn48605.2020.9206750

Cited by 67 publications

(81 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The Convolutional neural network (CNN) algorithm is used to extract additional features from the raw data of the EEG signal in several recent studies, for example: Ding et al [48] constructed a one-dimensional convolutional neural network (1D-CNN) to extract the features. This network consists of 3 sequential segments: the temporal learner consisting of an EEG input segment (4 EEG channels * 1024 data points for each channel * 1), the spatial learner consisting of (4 * 9), and the classifier [48].…”

Section: Deep Learning Methods (1d 2d and 3d-cnn)mentioning

confidence: 99%

A Comprehensive Review for Emotion Detection Based on EEG Signals: Challenges, Applications, and Open Issues

Abdulrahman¹,

Baykara²

2021

View full text Add to dashboard Cite

Emotion classification based on physiological signals has become a hot topic in the past decade. Many studies have attempted to classify emotions using various techniques, to discover human emotions accurately. This study focused on listing the most recent studies that have classified emotions based on electroencephalogram (EEG) signals. This study also focused on solving the problems and challenges facing researchers in emotion classification and EEG applications used in several fields. The plan of this study is based on a strategy with three aspects within specific rules: The first aspect is the methods; we chose studies that included new methods to extract features. The second aspect is the data sets. We tried to choose a study that classified the same data set. The third aspect is applications; we have listed many applications of the EEG in several areas. We concluded from this study that detecting human emotions using the EEG signals is one of the most reliable and widely used methods of detecting emotions in the past few years. Also, we have noticed that the EEG can detect human emotions, especially in psychiatry, for example, for epileptic patients whose emotions cannot be extracted using traditional methods such as facial expressions and tone of voice.

show abstract

Section: Deep Learning Methods (1d 2d and 3d-cnn)mentioning

confidence: 99%

A Comprehensive Review for Emotion Detection Based on EEG Signals: Challenges, Applications, and Open Issues

Abdulrahman¹,

Baykara²

2021

View full text Add to dashboard Cite

show abstract

“…Liang et al [14] proposed an unsupervised clustering method based on multiple types of EEG features to classify emotions. Compared with transitional machine learning methods, deep learning methods have shown promising results in BCI domain, such as motor imagery classification [15] [16] [17] [18], emotion recognition [19] [20] [21], and mental-task classification [22] [23]. Yang et al [24] designed a neural network with sub-network nodes for EEG emotion recognition.…”

Section: Introductionmentioning

confidence: 99%

“…Lawhern et al [18] proposed EEGNet, an end-to-end deep learning framework that uses three convolutional layers to extract the temporal and spatial patterns from EEG data. Ding et al [21] designed TSception that utilizes multi-scale convolutional kernels to learn discriminative representations from affective EEG signals.…”

Section: Introductionmentioning

confidence: 99%

“…To evaluate the proposed method, learning Local-Global-Graph representations (LGG), we conduct emotion classification experiments on the publicly available benchmark dataset, a Database for Emotion Analysis using Physiological signals (DEAP) [30]. The proposed LGG is compared with several state-of-the-art methods in the BCI domain, namely FBFgMDM [31], FBTSC [31], Unsupervised learning [14], DeepConvNet [15], ShallowConvNet [15], EEGNet [18], and TSception [21]. From the experiment results, LGG achieves significantly (p < 0.05) higher accuracy than other methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LGGNet: Learning from Local-Global-Graph Representations for Brain-Computer Interface

Ding¹,

Robinson²,

Zeng³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we propose LGG, a neurologically inspired graph neural network, to learn local-global-graph representations from Electroencephalography (EEG) for a Brain-Computer Interface (BCI). A temporal convolutional layer with multi-scale 1D convolutional kernels and kernel-level attention fusion is proposed to learn the temporal dynamics of EEG. Inspired by neurological knowledge of cognitive processes in the brain, we propose local and global graph-filtering layers to learn the brain activities within and between different functional areas of the brain to model the complex relations among them during the cognitive processes. Under the robust nested cross-validation settings, the proposed method is evaluated on the publicly available dataset DEAP, and the classification performance is compared with state-of-the-art methods, such as FBFgMDM, FBTSC, Unsupervised learning, DeepConvNet, ShallowConvNet, EEGNet, and TSception. The results show that the proposed method outperforms all these state-of-the-art methods, and the improvements are statistically significant (p < 0.05) in most cases. The source code can be found at: https://github.com/yi-ding-cs/LGG

show abstract

“…For image and signal data, good data pre-processing can effectively reduce noise data, thus effectively improving accuracy [8]. The emergence of TScpetion [17]solves the above problems, which effectively combines features in the time-domain and frequency-domain. However, it is not perfect in global feature extraction and data pre-processing, whose recognition accuracy needs to be further improved.…”

Section: Introductionmentioning

confidence: 99%

GTSception: A Deep Learning EEG Emotion Recognition Model Based on Fusion of Global, Time Domain and Frequency Domain Feature Extraction

Jian

Zhang

Qiu

et al. 2021

Preprint

View full text Add to dashboard Cite

With the rapid development of deep learning in recent years, automatic electroencephalography (EEG) emotion recognition has been widely concerned. At present, most deep learning methods do not normalize EEG data properly and do not fully extract the features of time and frequency domain, which will affect the accuracy of EEG emotion recognition. To solve these problems, we propose GTScepeion, a deep learning EEG emotion recognition model. In pre-processing, the EEG time slicing data including channels were pre-processed. In our model, global convolution kernels are used to extract overall semantic features, followed by three kinds of temporal convolution kernels representing different emotional periods, followed by two kinds of spatial convolution kernels highlighting brain hemispheric differences to extract spatial features, and finally emotions are dichotomy classified by the full connected layer. The experiments is based on the DEAP dataset, and our model can effectively normalize the data and fully extract features. For Arousal, ours is 8.76% higher than the current optimal emotion recognition model based on Inception. For Valence, the best accuracy of our model reaches 91.51%.

show abstract

TSception:A Deep Learning Framework for Emotion Detection Using EEG

Cited by 67 publications

References 34 publications

A Comprehensive Review for Emotion Detection Based on EEG Signals: Challenges, Applications, and Open Issues

A Comprehensive Review for Emotion Detection Based on EEG Signals: Challenges, Applications, and Open Issues

LGGNet: Learning from Local-Global-Graph Representations for Brain-Computer Interface

GTSception: A Deep Learning EEG Emotion Recognition Model Based on Fusion of Global, Time Domain and Frequency Domain Feature Extraction

Contact Info

Product

Resources

About