Xuyun Wen

Xuyun Wen

Auto-generated profile, based on publicly available information

Xuyun Wen

Nanjing University of Aeronautics and Astronautics

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 May 2024 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
PGPeiliang Gong
Peiliang Gong
PWPengpai Wang
Pengpai Wang
YZYueying Zhou
Yueying Zhou
XWXuyun Wen
Xuyun Wen
DZDaoqiang Zhang
Daoqiang Zhang
Fatigue driving is a significant cause of road traffic accidents and associated casualties. Automatic assessment of driver drowsiness by monitoring electroencephalography (EEG) signals offer a more objective way to improve driving safety. However, most existing measures are based on multi-channel EEG signals, which are more difficult to apply in practical scenarios as it usually lacks better portability and comfort. In addition, due to the relatively parsimonious and non-stationary characteristics, it is still challenging to effectively accomplish drowsiness recognition by exploiting single-channel EEG signals alone. To this end, we propose a novel temporal-frequential attentional convolutional neural network (TFAC-Net) to take full advantage of spectral-temporal features for single-channel EEG driver drowsiness recognition. Specifically, to capture the potentially valuable information contained in single-channel EEG, the continuous wavelet transform is first employed to generate a corresponding spectral-temporal representation. Then, the temporal-frequential attention mechanism is adopted to reveal critical time-frequency regions in terms of the driver’s mental state. Finally, an adaptive feature fusion module is considered to recalibrate and integrate the most relevant feature channels for final prediction. Extensive experimental results on a widely used public EEG driving dataset demonstrate that the TFAC-Net approach is superior to the state-of-the-art methods, and could discover some discriminative temporal-frequential regions. Moreover, this study also sheds light on the development of portable EEG devices and practical driver drowsiness recognition.

26 June 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YZYueying Zhou
Yueying Zhou
PWPengpai Wang
Pengpai Wang
PGPeiliang Gong
Peiliang Gong
FWFulin Wei
Fulin Wei
XWXuyun Wen
Xuyun Wen
DZDaoqiang Zhang
Daoqiang Zhang
Regarding cognitive workload recognition, electroencephalography (EEG) signals are nonstationary across time and vary from different subjects, thus hindering the cross-subject recognition performance. Though subject calibration or collecting massive training data may ease the above problem, it is generally time-consuming and expensive. In this paper, we propose a Deep Domain Adaptation (DDA) scheme for EEG-based cross-subject cognitive workload recognition, using the knowledge from the existing subjects (source domain) to improve the recognition performance of a new subject (target domain). Precisely, the proposed DDA method composes four modules, EEG features extractor, label classifier, feature distribution alignment, and domain discriminator. The model starts with the EEG feature extractor to learn the shallow feature representation for both domains. The label classifier further learns the deep representation and trains the classifier supervised. Finally, the feature distribution alignment matches the shallow feature distribution discrepancy, and the domain discriminator matches the deep distribution discrepancy by the adversarial training with the feature extractor. It not only learns domain-invariant features but also achieves robust domain-adaptive cross-subject recognition results in an end-to-end training framework. We conduct experiments to classify low and high workloads on a self-designed EEG dataset and one public EEG dataset. Experimental results demonstrate that our DDA scheme significantly outperforms the baselines and other state-of-the-art methods with improvements of 2% to 9% in terms of recognition accuracy.

1 October 2020 in Cerebral Cortex
Verified
Journal Article
Open access
PDF file
XWXuyun Wen
Xuyun Wen
RWRifeng Wang
Rifeng Wang
WYWeiyan Yin
Weiyan Yin
WLWeili Lin
Weili Lin
HZHan Zhang
Han Zhang
DSDinggang Shen
Dinggang Shen
Uncovering the moment-to-moment dynamics of functional connectivity (FC) in the human brain during early development is crucial for understanding emerging complex cognitive functions and behaviors. To this end, this paper leveraged a longitudinal resting-state functional magnetic resonance imaging dataset from 51 typically developing infants and, for the first time, thoroughly investigated how the temporal variability of the FC architecture develops at the “global” (entire brain), “mesoscale” (functional system), and “local” (brain region) levels in the first 2 years of age. Our results revealed that, in such a pivotal stage, 1) the whole-brain FC dynamic is linearly increased; 2) the high-order functional systems tend to display increased FC dynamics for both within- and between-network connections, while the primary systems show the opposite trajectories; and 3) many frontal regions have increasing FC dynamics despite large heterogeneity in developmental trajectories and velocities. All these findings indicate that the brain is gradually reconfigured toward a more flexible, dynamic, and adaptive system with globally increasing but locally heterogeneous trajectories in the first 2 postnatal years, explaining why infants have rapidly developing high-order cognitive functions and complex behaviors.

28 November 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
PWPengpai Wang
Pengpai Wang
PGPeiliang Gong
Peiliang Gong
YZYueying Zhou
Yueying Zhou
XWXuyun Wen
Xuyun Wen
DZDaoqiang Zhang
Daoqiang Zhang
In the field of brain-computer interface (BCI), brain decoding using EEG is an essential direction, motion imagery EEG-based BCI can not only help rehabilitation of patients with physical disabilities, but also enhance the endurance and power of people. Most of the existing MI-based BCI studies are limited to discrete EEG classification or 3D directional limb trajectory reconstruction. To suitable for the requirements of BCI systems in practical applications, here we explored the decoding of trajectories of continuous non-directional motion imagination in three-dimensional (3D) space based on Chinese Sign Language. We propose Motion Imagery Trajectory Reconstruction Transformer (MITRT) model to decode the EEG signals of the subjects performing motion imagery, and obtain the positional changes in the 3D space of the shoulder, elbow, and wrist skeleton points in the neural activity. We add the geometric constraint features of upper limb skeleton points to the model, the MITRT decoding model can obtain prior knowledge to improve the reconstruction accuracy of spatial positions. To verify the decoding performance of our proposed model, we collected motor imagery EEG signals of 20 subjects based on Chinese Sign Language for experiments. The experimental results show that the average Pearson correlation coefficient of the six skeleton points was 0.975, which was significantly higher than the contrast models. This study is the first attempt to reconstruct multi-directional continuous non-directional upper limb motor imagery trajectories based on Chinese Sign Language. The experimental results show that it is feasible to decode and reconstruct imagined 3D trajectories of human upper limb skeleton points from scalp EEG.

22 January 2024 in IEEE Transactions on Neural Systems and Rehabilitation Engineering
Verified
Journal Article
Open access
XWXuyun Wen
Xuyun Wen
QCQumei Cao
Qumei Cao
BJBin Jing
Bin Jing
DZDaoqiang Zhang
Daoqiang Zhang
Predicting individual behavior from brain imaging data using machine learning is a rapidly growing field in neuroscience. Functional connectivity (FC), which captures interactions between different brain regions, contains valuable information about the organization of the brain and is considered a crucial feature for modeling human behavior. Graph convolutional networks (GCN) have proven to be a powerful tool for extracting graph structure features and have shown promising results in various FC-based classification tasks, such as disease classification and prognosis prediction. Despite this success, few behavior prediction models currently exist based on GCN, and their performance is not satisfactory. To address this gap, a new model called the Multi-Scale FC-based Multi-Order GCN (MSFC-MO-GCN) was proposed in this paper. The model considers the hierarchical structure of the brain system and utilizes FCs inferred from multiple spatial scales as input to comprehensively characterize individual brain organization. To enhance the feature learning ability of GCN, a multi-order graph convolutional layer is incorporated, which uses multi-order neighbors to guide message passing and learns high-order graph information of nodal connections. Additionally, an inter-subject contrast constraint is designed to control the potential information redundancy of FCs among different spatial scales during the feature learning process. Experimental evaluation were conducted on the publicly available dataset from human connectome project. A total of 805 healthy subjects were included and 5 representative behavior metrics were used. The experimental results show that our proposed method outperforms the existing behavior prediction models in all behavior prediction tasks.

2 April 2025 in Human Brain Mapping
Verified
Journal Article
Open access
XWXiaoya Wu
Xiaoya Wu
CLChuang Liang
Chuang Liang
JBJuan Bustillo
Juan Bustillo
PKPeter Kochunov
Peter Kochunov
XWXuyun Wen
Xuyun Wen
SQShile Qi
Shile Qi
Neuropsychiatric disorders are associated with altered functional connectivity (FC); however, the reported regional patterns of functional alterations suffered from low replicability and high variability. This is partly because of differences in the atlas and delineation techniques used to measure FC‐related deficits within/across disorders. We systematically investigated the impact of the brain parcellation approach on the FC‐based brain network analysis. We focused on identifying the replicable FCs using three structural brain atlases, including Automated Anatomical Labeling (AAL), Brainnetome atlas (BNA) and HCP_MMP_1.0, and four functional brain parcellation approaches: Yeo‐Networks (Yeo), Gordon parcel (Gordon) and two Schaefer parcelletions, among correlation, group difference, and classification tasks in six neuropsychiatric disorders: attention deficit and hyperactivity disorder (ADHD, n = 340), autism spectrum disorder (ASD, n = 513), schizophrenia (SZ, n = 200), schizoaffective disorder (SAD, n = 142), bipolar disorder (BP, n = 172), and major depression disorder (MDD, n = 282). Our cross‐atlas/disorder analyses demonstrated that frontal‐related FC deficits were reproducible in all disorders, independent of the atlasing approach; however, replicable FC extraction in other areas and the classification accuracy were affected by the parcellation schema. Overall, functional atlases with finer granularity performed better in classification tasks. Specifically, the Schaefer atlases generated the most repeatable FC deficit patterns across six illnesses. These results indicate that frontal‐related FCs may serve as potential common and robust neuro‐abnormalities across 6 psychiatric disorders. Furthermore, in order to improve the replicability of rsfMRI‐based FC analyses, this study suggests the use of functional templates at larger granularity.