PCPeiyin Chen

Peiyin Chen

*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


Peiyin Chen has a total of 43 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • End‐to‐End EEG ‐Based Schizophrenia Detection via Hybrid Temporal–Spatial Dependency Learning

    by Peiyin Chen, Xin Yang, Han Zhang, Yichen Huang, Hui Feng, Yu Wang

    Electroencephalogram (EEG)‐based schizophrenia (SZ) detection is promising because EEG provides a noninvasive characterisation of abnormal brain activity. However, many existing methods still depend on handcrafted features or insufficiently model the coupled temporal–spatial dependencies in EEG signals. Moreover, a considerable number of prior studies are evaluated on only a single dataset, leaving their robustness across heterogeneous acquisition conditions unclear. To address these issues, we propose a novel end‐to‐end framework, termed Hybrid Temporal–Spatial Dependency Learning Network (HTSD), for automatic SZ detection directly from raw EEG signals. HTSD integrates convolution‐based temporal encoding and transformer‐based spatial dependency learning in a unified architecture. Specifically, a multibranch local temporal encoder is designed to capture diverse short‐range temporal dynamics, while a spatial correlation encoder based on multihead self‐attention explicitly models adaptive interchannel dependencies and long‐range interactions among brain regions. A global temporal encoder is further employed to aggregate high‐level temporal representations for classification. We evaluate HTSD on three public EEG datasets under both segment‐level five‐fold cross‐validation and subject‐level leave‐one‐subject‐out (LOSO) cross‐subject validation. HTSD achieves five‐fold accuracies of 98.03%, 99.51% and 97.12% and LOSO accuracies of 93.87%, 96.43% and 91.74% on the three datasets, respectively. These results indicate that HTSD provides accurate SZ detection while retaining promising robustness under subject‐independent evaluation.

  • Open Access

    LVLMs as inspectors: An agentic framework for category-level structural defect annotation

    by Sheng Jiang, Yuanmin Ning, Bingxi Huang, Weiyu Cheng, Peiyin Chen, Zhaohui Chen

  • Exploring Grounding Abilities in Vision-Language Models Through Contextual Perception

    by Wei Xu, Tianfei Zhou, Taoyuan Zhang, Jie Li, Peiyin Chen, Jia Pan, Xiaofeng Liu

    Vision language models (VLMs) have demonstrated strong general capabilities and achieved great success in areas such as image understanding and reasoning. Visual prompts enhance the focus of VLMs on designated areas, but their fine-grained grounding has not been fully developed. Recent research has used Set-of-Mark (SoM) approach to unleash the grounding capabilities of Generative Pre-trained Transformer-4 with Vision (GPT-4V), achieving significant benchmark performance. However, SoM still has problems with label offset and hallucination of vision language models, and the grounding ability of VLMs remains limited, making it challenging to handle complex scenarios in human-robot interaction. To address these limitations and provide more accurate and less hallucinatory results, we propose Contextual Set-of-Mark (ConSoM), a new SoM-based prompting mechanism that leverages dual-image inputs and contextual semantic information of images. Experiments demonstrate that ConSoM has distinct advantages in visual grounding, improving by 11% compared to the baseline on the dataset Refcocog. Furthermore, we evaluated ConSoM’s grounding abilities in five indoor scenarios, where it exhibited strong robustness in complex environments and under occlusion conditions. We also introduced a scalable annotation method for pixel-level question-answering dataset. The accuracy, scalability, and depth of world knowledge make ConSoM a highly effective approach for future human-robot interactions.


  • Unsupervised Domain Adaptation With Synchronized Self-Training for Cross- Domain Motor Imagery Recognition

    by Peiyin Chen, Xiaofeng Liu, Chao Ma, He Wang, Xiong Yang, Celso Grebogi, Xiao Gu, Zhongke Gao

    Robust decoding performance is essential for the practical deployment of brain-computer interface (BCI) systems. Existing EEG decoding models often rely on large amounts of annotated data collected through specific experimental setups, which fail to address the heterogeneity of data distributions across different domains. This limitation hinders BCI systems from effectively managing the complexity and variability of real-world data. To overcome these challenges, we propose Synchronized Self-Training Domain Adaptation (SSTDA) for cross-domain motor imagery classification. Specifically, SSTDA leverages labeled signals from a source domain and applies self-training to unlabeled signals from a target domain, enabling the simultaneous training of a more robust classifier. The raw EEG signals are mapped into a latent space by a feature extractor for discriminative representation learning. A domain-shared latent space is then learned by optimizing the feature extractor with both source and target samples, using an easy-tohard self-training process. We validate the method with extensive experiments on two public motor imagery datasets: Dataset IIa of BCI Competition IV and the High Gamma dataset. In the inter-subject task, our method achieves classification accuracies of 64.43% and 80.40%, respectively. It also outperforms existing methods in the inter-session task. Moreover, we develope a new six-class motor imagery dataset and achieve test accuracies of 77.09% and 80.18% across different datasets. All experimental results demonstrate that our SSTDA outperforms existing algorithms in inter-session, inter-subject, and inter-dataset validation protocols, highlighting its capability to learn discriminative, domain-invariant representations that enhance EEG decoding performance.


  • Co-Attention-Based Cross-Stitch Network for Parameter Prediction of Two-Phase Flow

    by Jianbo Zhang, Chao Ma, Peiyin Chen, Mengyu Li, Ruiqi Wang, Zhongke Gao

    Gas-liquid two-phase flow is pervasive in industrial processes, and accurate measurement of its parameters can effectively improve industrial efficiency. Void fraction, a key parameter in gas-liquid two-phase flow, plays a crucial role in mixture density determination and flow structure analysis. In this paper, a novel multi-task learning method is proposed to predict the gas void fraction. Firstly, we conduct vertical upward gas-liquid two-phase flow experiments to obtain fluid signals relying on the four-sector distributed conductance sensor. Then, we pair the measurement of gas void fraction and its classification and design a Co-Attention based Cross-Stitch Network (CACSNet) to process these two closely related tasks simultaneously. In CACSNet, two parallel embedding branches are adopted to extract task-specific features. There are three feature extractors in each branch to capture in-depth representations of temporal and spatial. To learn an optimal combination of these task-specific features and generate shared representations, we propose the Cross-stitch Embedding with Attention Modules (CEAM), which is incorporated the Channel Co-Attention Module (CCAM) and the Temporal Co-Attention Module (TCAM). Finally, we compare our CACSNet with single-task baseline and other competitive methods, which proves that our framework achieves superior performance in gas void fraction prediction. We also conduct detailed model ablation and parameter analysis to illustrate the efficiency of our proposed structure.


  • EEG-Based Motor Imagery Recognition Framework via Multisubject Dynamic Transfer and Iterative Self-Training

    by He Wang, Peiyin Chen, Meng Zhang, Jianbo Zhang, Xinlin Sun, Mengyu Li, Xiong Yang, Zhongke Gao

    A robust decoding model that can efficiently deal with the subject and period variation is urgently needed to apply the brain–computer interface (BCI) system. The performance of most electroencephalogram (EEG) decoding models depends on the characteristics of specific subjects and periods, which require calibration and training with annotated data prior to application. However, this situation will become unacceptable as it would be difficult for subjects to collect data for an extended period, especially in the rehabilitation process of disability based on motor imagery (MI). To address this issue, we propose an unsupervised domain adaptation framework called iterative self-training multisubject domain adaptation (ISMDA) that focuses on the offline MI task. First, the feature extractor is purposefully designed to map the EEG to a latent space of discriminative representations. Second, the attention module based on dynamic transfer matches the source domain and target domain samples with a higher coincidence degree in latent space. Then, an independent classifier oriented to the target domain is employed in the first stage of the iterative training process to cluster the samples of the target domain through similarity. Finally, a pseudolabel algorithm based on certainty and confidence is employed in the second stage of the iterative training process to adequately calibrate the error between prediction and empirical probabilities. To evaluate the effectiveness of the model, extensive testing has been performed on three publicly available MI datasets, the BCI IV IIa, the High gamma dataset, and Kwon et al. datasets. The proposed method achieved 69.51%, 82.38%, and 90.98% cross-subject classification accuracy on the three datasets, which outperforms the current state-of-the-art offline algorithms. Meanwhile, all results demonstrated that the proposed method could address the main challenges of the offline MI paradigm.


  • A Novel Complex Network-Based Graph Convolutional Network in Major Depressive Disorder Detection

    by Xinlin Sun, Chao Ma, Peiyin Chen, Mengyu Li, He Wang, Weidong Dang, Chaoxu Mu, Zhongke Gao

    As a worldwide disease, major depressive disorder (MDD) severely damages patients’ mental health. It is of great significance of detecting MDD accurately in providing necessary guidance for physicians. Here, a novel complex network-based graph convolutional network (CN-GCN), is developed to detect MDD. First, multichannel electroencephalogram (EEG) signals are decomposed into several frequency bands. Then, a multilayer brain network is constructed via a phase-locking value (PLV), where each layer corresponds to a specific frequency band. Aiming at accurately identifying brain states, the CN-GCN is developed, with multilayer brain network as input. Moreover, power spectral density (PSD) is applied for refining node-level rhythm features. Such structure of CN-GCN allows learning the node features based on the topology connections of the brain network. The proposed framework shows the state-of-the-art (SOTA) detection accuracy of 99.29% on a public MDD dataset. Our work confirms the validity on integrating complex network and GCN in multichannel EEG signal analysis and contributes to identifying complex brain states better.


  • Open Access

    Transfer Learning With Optimal Transportation and Frequency Mixup for EEG-Based Motor Imagery Recognition

    by Peiyin Chen, He Wang, Xinlin Sun, Haoyu Li, Celso Grebogi, Zhongke Gao

    Electroencephalography-based Brain Computer Interfaces (BCIs) invariably have a degenerate performance due to the considerable individual variability. To address this problem, we develop a novel domain adaptation method with optimal transport and frequency mixup for cross-subject transfer learning in motor imagery BCIs. Specifically, the preprocessed EEG signals from source and target domain are mapped into latent space with an embedding module, where the representation distributions and label distributions across domains have a large discrepancy. We assume that there exists a non-linear coupling matrix between both domains, which can be utilized to estimate the distance of joint distributions for different domains. Depending on the optimal transport, the Wasserstein distance between source and target domains is minimized, yielding the alignment of joint distributions. Moreover, a new mixup strategy is also introduced to generalize the model, where the inputs trials are mixed in frequency domain rather than in raw space. The extensive experiments on three evaluation benchmarks are conducted to validate the proposed framework. All the results demonstrate that our method achieves a superior performance than previous state-of-the-art domain adaptation approaches.


  • A Multifeatured Time–Frequency Neural Network System for Classifying sEMG

    by Zhongke Gao, Yangyang Wang, Xinlin Sun, Peiyin Chen, Chao Ma

    Human gesture-related electromyographic signal in intelligent recognition systems has attracted widespread attention. The EMG signal is the temporal and spatial superposition of motor unit neural potentials and it contains nonlinear and complex characteristics. Despite the previous efforts, recognizing more different actions and building an online recognition system remain challenging. To address these issues, we construct a multi-feature time-frequency neural network system and a dataset which contains twenty kinds of hand movements. In a multi-featured time-frequency neural network system, the multi-layer CNN structure is used to obtain faster recognition and better classification results. The experimental results demonstrate that equipment and algorithmic model can reach a 94.66% accuracy. Our approach focuses on processing multiple time and frequency domain features of sEMG using CNN. Importantly, we also perform real-time validation. The system can accurately recognize hand movements to control UAV motion in real-time, in which case system classification takes 0.05s. This demonstrates the practicality and effectiveness of the sEMG acquisition system based on the multifeatured time-frequency neural network.