FHFazheng Hou

Fazheng Hou

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Fazheng Hou has a total of 33 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    EAMCCA: Euclidean Alignment based Multiscale Convolution and Channel Attention Network for MI-EEG Decoding

    by Wei Meng, Jiahao Wu, Fazheng Hou, Jie Zuo, Li Ma, Quan Liu

    Electroencephalography (EEG)-based motor imagery (MI), as an important paradigm of brain computer interfaces (BCIs), has been explored for applications in fields like medical rehabilitation, driven by the continuous advancement of sensor and artificial intelligence technologies. However, it is still challenging to effectively utilize deep learning to obtain the most discriminative MI related EEG features. In this paper, we propose Euclidean alignment based multiscale convolution and channel attention (EAMCCA) network for MI-EEG decoding. Following EA preprocessing of raw EEG signals, the network employs two parallel branches to extract complementary features: the multiscale convolutional module that captures spatial-temporal features across different scales and the channel attention module that prioritizes features from MI-strongly relevant channels. Moreover, it utilizes the multi-head attention module, to capture the features’ contextual relationships and global temporal correlations. Finally, by capturing the two deep feature representations and modeling their interdependencies, the convolution module fuses them into highly discriminative features, which substantially enhances the performance of MI classification. We validated EAMCCA on three datasets: the BCI Competition IV-2a dataset, the BCI Competition IV-2b dataset, and the High Gamma Dataset (HGD). The experimental results indicate that EAMCCA achieves state-of the-art performance, with accuracies of 86.61%, 89.42%, and 93.61% in the respective experiments. The proposed method provides a strong reference for future personalized MI-EEG decoding with applications in rehabilitation.

  • Visually-Inspired Multimodal Iterative Attentional Network for High-Precision EEG—Eye-Movement Emotion Recognition

    by Wei Meng, Fazheng Hou, Kun Chen, Li Ma, Quan Liu

    Advancements in artificial intelligence have propelled affective computing toward unprecedented accuracy and real-world impact. By leveraging the unique strengths of brain signals and ocular dynamics, we introduce a novel multimodal framework that integrates EEG and eye-movement (EM) features synergistically to achieve more reliable emotion recognition. First, our EEG Feature Encoder (EFE) uses a convolutional architecture inspired by the human visual cortex’s eccentricity–receptive-field mapping, enabling the extraction of highly discriminative neural patterns. Second, our EM Feature Encoder (EMFE) employs a Kolmogorov–Arnold Network (KAN) to overcome the sparse sampling and dimensional mismatch inherent in EM data; through a tailored multilayer design and interpolation alignment, it generates rich, modality-compatible representations. Finally, the core Multimodal Iterative Attentional Feature Fusion (MIAFF) module unites these streams: alternating global and local attention via a Hierarchical Channel Attention Module (HCAM) to iteratively refine and integrate features. Comprehensive evaluations on SEED (3-class) and SEED-IV (4-class) benchmarks show that our method reaches leading-edge accuracy. However, our experiments are limited by small homogeneous datasets, untested cross-cultural robustness, and potential degradation in noisy or edge-deployment settings. Nevertheless, this work not only underscores the power of biomimetic encoding and iterative attention but also paves the way for next-generation brain–computer interface applications in affective health, adaptive gaming, and beyond.


  • Emotion recognition via affective EEG signals: State of the art

    by Wei Meng, Fazheng Hou, Mengyuan Zhao, Jingjing Kong, Jiahao Wu, Jie Zuo, Quan Liu


  • MECAM: A Novel Multi-Axis EEG Channel Attention Model for Emotion Recognition

    by Fazheng Hou, Wei Meng, Li Ma, Kun Chen, Qingsong Ai, Quan Liu, Sheng Q. Xie

    With the constant evolution of information technology, the paradigm of human-computer interaction has progressively transitioned from emphasizing command behavior to a novel approach that prioritizes natural language and emotional communication. In response to this trend, the development of an effective model for emotion recognition has become imperative. Consequently, this article introduces a novel multi-axis EEG channel attention model (MECAM) designed for emotion recognition. To enhance the model’s capabilities, we employ a split depth-wise convolution with a larger convolutional kernel, resulting in the creation of four parallel branches. This not only reduces the computational complexity but also expands the receptive field of the model. The incorporation of a multi-axis attention mechanism, capturing both global and local information from the output features of parallel branches, further elevates the network’s receptive field through the MBConv block. The primary objective of MECAM is to adeptly extract discriminative features from electroencephalogram (EEG) data, thereby enhancing the performance of emotion recognition. The model underwent rigorous validation using multiple datasets, including database for emission analysis using physical signals (DEAPs), SJTU emission EEG dataset (SEED), and SJTU emission EEG dataset with four emission (SEED-IV). In subject-dependent experiments, the accuracy of binary and ternary emotion classification tasks exceeded 95%, while the accuracy of the quaternary classification task surpassed 93%. In subject-independent experiments, the accuracy of most emotion classification tasks also exceeded 90%, with the exception of quaternary classification on DEAP, achieving an accuracy of 86.64%. The experimental results unequivocally underscore the superior performance of MECAM in EEG-based emotion recognition tasks.


  • Deep Feature Extraction and Attention Fusion for Multimodal Emotion Recognition

    by Zhiyi Yang, Dahua Li, Fazheng Hou, Yu Song, Qiang Gao

    Recently, electroencephalogram (EEG)-based multimodal emotion recognition has emerged as one of the research hotspots in affective computing. However, the existing methods tend to ignore the interaction information between the EEG and other modal features. In this brief, we propose a novel model termed EEANet (EEG and eye movement Attention Network) to find the modal correlation at feature level. The DE feature and 31 eye movement features were extracted from the pre-processed EEG and eye movement signals, and then two feedforward encoders were used to capture the deep features, respectively. The interactive attention layer is applied to learn multi-modal complementary information and semantic-level context information. Finally, the multi-head self-attention mechanism allows the model to focus on the discriminative features for emotion classification. The model was verified on the SEED-IV dataset, and the results showed that the accuracy of emotion recognition was significantly improved with the EEANet, and the average accuracy of the four classifications was 92.26%.


  • SECT: A Method of Shifted EEG Channel Transformer for Emotion Recognition

    by Zhongli Bai, Fazheng Hou, Kaixuan Sun, Qingzhou Wu, Mu Zhu, Zemin Mao, Yu Song, Qiang Gao

    Recently, electroencephalographic (EEG) emotion recognition attract attention in the field of human-computer interaction (HCI). However, most of the existing EEG emotion datasets primarily consist of data from normal human subjects. To enhance diversity, this study aims to collect EEG signals from 30 hearing-impaired subjects while they watch video clips displaying six different emotions (happiness, inspiration, neutral, anger, fear, and sadness). The frequency domain feature matrix of EEG signals, which comprise power spectral density (PSD) and differential entropy (DE), were up-sampled using cubic spline interpolation to capture the correlation among different channels. To select emotion representation information from both global and localized brain regions, a novel method called Shifted EEG Channel Transformer (SECT) was proposed. The SECT method consists of two layers: the first layer utilizes the traditional channel Transformer (CT) structure to process information from global brain regions, while the second layer acquires localized information from centrally symmetrical and reorganized brain regions by shifted channel Transformer (S-CT). We conducted a subject-dependent experiment, and the accuracy of the PSD and DE features reached 82.51% and 84.76%, respectively, for the six kinds of emotion classification. Moreover, subject-independent experiments were conducted on a public dataset, yielding accuracies of 85.43% (3-classification, SEED), 66.83% (2-classification on Valence, DEAP), and 65.31% (2-classification on Arouse, DEAP), respectively.


  • EEG-Based Emotion Recognition for Hearing Impaired and Normal Individuals With Residual Feature Pyramids Network Based on Time–Frequency–Spatial Features

    by Fazheng Hou, Junjie Liu, Zhongli Bai, Zhiyi Yang, Jiayin Liu, Qiang Gao, Yu Song

    With the development of affective computing, discriminative feature selection is critical for electroencephalography (EEG) emotion recognition. In this paper, we fused four EEG feature matrices constructed by the preprocessed signal, the differential entropy, the symmetric difference, and the symmetric quotient based on the International 10-20 system, which integrates time domain, frequency domain, and spatial domain information of EEG signals. For the feature classification model, we used the Space-to-Depth (S2D) layer instead of the convolutional neural network (CNN) as the backbone to reduce the calculation of the model without affecting the classification performance. The Residual Feature Pyramid Network (RFPN) was proposed to obtain the correlation of channels, and then the deep multi-scale semantic information of EEG feature maps is captured. The emotion classification strategy was evaluated by DEAP, SEED, SEED-IV and our Hearing-Impaired EEG Dataset (HIED). The classification accuracies were 93.56% (4-class, DEAP), 96.84% (3-class, SEED), 91.62% (4-class, SEED-IV), and 87.74% (6-class, HIED), respectively. Furthermore, we also found that the difference in emotional response between the left and right brain regions of hearing-impaired subjects is more obvious than that of normal subjects.


  • Open Access

    Emotion Recognition of Subjects With Hearing Impairment Based on Fusion of Facial Expression and EEG Topographic Map

    by Dahua Li, Jiayin Liu, Yi Yang, Fazheng Hou, Haotian Song, Yu Song, Qiang Gao, Zemin Mao

    Emotion analysis has been employed in many fields such as human-computer interaction, rehabilitation, and neuroscience. But most emotion analysis methods mainly focus on healthy controls or depression patients. This paper aims to classify the emotional expressions in individuals with hearing impairment based on EEG signals and facial expressions. Two kinds of signals were collected simultaneously when the subjects watched affective video clips, and we labeled the video clips with discrete emotional states (fear, happiness, calmness, and sadness). We extracted the differential entropy (DE) features based on EEG signals and converted DE features into EEG topographic maps (ETM). Next, the ETM and facial expressions were fused by the multichannel fusion method. Finally, a deep learning classifier CBAM_ResNet34 combined Residual Network (ResNet) and Convolutional Block Attention Module (CBAM) was used for subject-dependent emotion classification. The results show that the average classification accuracy of four emotions recognition after multimodal fusion achieves 78.32%, which is higher than 67.90% for facial expressions and 69.43% for EEG signals. Moreover, visualization by the Gradient-weighted Class Activation Mapping (Grad-CAM) of ETM showed that the prefrontal, temporal and occipital lobes were the brain regions closely related to emotional changes in individuals with hearing impairment.


  • A Novel Domain Adversarial Networks Based on 3D-LSTM and Local Domain Discriminator for Hearing-Impaired Emotion Recognition

    by Zekun Tian, Dahua Li, Yi Yang, Fazheng Hou, Zhiyi Yang, Yu Song, Qiang Gao

    Recent research on emotion recognition suggests that deep network-based adversarial learning has an ability to solve the cross-subject problem of emotion recognition. This study constructed a hearing-impaired electroencephalography (EEG) emotion dataset containing three emotions (positive, neutral, and negative) in 15 subjects. The emotional domain adversarial neural network (EDANN) was carried out to identify hearing-impaired subjects' emotions by learning hidden emotion information between the labeled data and the data with no-label. For the input data, we propose a spatial filter matrix to reduce the overfitting of the training data. A feature extraction network 3DLSTM-ConvNET was used to extract comprehensive emotional information from the time, frequency, and spatial dimensions. Moreover, emotion local domain discriminator and emotion film group local domain discriminator were added to reduce the distribution distance between the same kinds of emotions and different film groups, respectively. According to the experimental results, the average accuracy of subject-dependent is 0.984 (STD: 0.011), and that of subject-independent is 0.679 (STD: 0.140). In addition, by analyzing the discrimination characteristics, we found that the brain regions with emotional recognition in the hearing-impaired are distributed in the wider areas of the parietal and occipital lobes, which may be caused by visual processing.