Xin Wei

Dr. Xin Wei

orcid

Research Focus & Metrics

Representative Publications

View All Publications
30 September 2022 in Frontiers in Bioengineering and Biotechnology
Verified
Journal Article
Open access
XLXiaoyin Liu
Xiaoyin Liu
JWJingjing Wang
Jingjing Wang
PWPeng Wang
Peng Wang
LZLin Zhong
Lin Zhong
SWShan Wang
Shan Wang
LZLiangxue Zhou
Liangxue Zhou
Regenerating brain defects after traumatic brain injury (TBI) still remains a significant difficulty, which has motivated interest in 3D printing to design superior replacements for brain implantation. Collagen has been applied to deliver cells or certain neurotrophic factors for neuroregeneration. However, its fast degradation rate and poor mechanical strength prevent it from being an excellent implant material after TBI. In the present study, we prepared 3D-printed collagen/silk fibroin/hypoxia-pretreated human umbilical cord mesenchymal stem cells (HUCMSCs)-derived exosomes scaffolds (3D-CS-HMExos), which possessed favorable physical properties suitable biocompatibility and biodegradability and were attractive candidates for TBI treatment. Furthermore, inspired by exosomal alterations resulting from cells in different external microenvironments, exosomes were engineered through hypoxia stimulation of mesenchymal stem cells and were proposed as an alternative therapy for promoting neuroregeneration after TBI. We designed hypoxia-preconditioned (Hypo) exosomes derived from HUCMSCs (Hypo-MExos) and proposed them as a selective therapy to promote neuroregeneration after TBI. For the current study, 3D-CS-HMExos were prepared for implantation into the injured brains of beagle dogs. The addition of hypoxia-induced exosomes further exhibited better biocompatibility and neuroregeneration ability. Our results revealed that 3D-CS-HMExos could significantly promote neuroregeneration and angiogenesis due to the doping of hypoxia-induced exosomes. In addition, the 3D-CS-HMExos further inhibited nerve cell apoptosis and proinflammatory factor (TNF-α and IL-6) expression and promoted the expression of an anti-inflammatory factor (IL-10), ultimately enhancing the motor functional recovery of TBI. We proposed that the 3D-CS-loaded encapsulated hypoxia-induced exosomes allowed an adaptable environment for neuroregeneration, inhibition of inflammatory factors and promotion of motor function recovery in TBI beagle dogs. These beneficial effects implied that 3D-CS-HMExos implants could serve as a favorable strategy for defect cavity repair after TBI.

14 August 2023 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
JZJianwei Zhang
Jianwei Zhang
LZLei Zhang
Lei Zhang
JWJunyou Wang
Junyou Wang
XWXin Wei
Xin Wei
JLJiaqi Li
Jiaqi Li
DDDan Du
Dan Du
Automated detection of skin lesions offers excellent potential for interpretative diagnosis and precise treatment of acne vulgar. However, the blurry boundary and small size of lesions make it challenging to detect acne lesions with traditional object detection methods. To better understand the acne detection task, we construct a new benchmark dataset named AcneSCU, consisting of 276 facial images with 31777 instance-level annotations from clinical dermatology. To the best of our knowledge, AcneSCU is the first acne dataset with high-resolution imageries, precise annotations, and fine-grained lesion categories, which enables the comprehensive study of acne detection. More importantly, we propose a novel method called Spatial Aware Region Proposal Network (SA-RPN) to improve the proposal quality of two-stage detection methods. Specifically, the representation learning for the classification and localization task is disentangled with a double head component to promote the proposals for hard samples. Then, Normalized Wasserstein Distance of each proposal is predicted to improve the correlation between the classification scores and the proposals' intersection-over-unions (IoUs). SA-RPN can serve as a plug-and-play module to enhance standard two-stage detectors. Extensive experiments are conducted on both AcneSCU and the public dataset ACNE04, and the results show that the proposed method can consistently outperform state-of-the-art methods. Code and the collected dataset are made available at https://github.com/pingguokiller/acnedetection to stimulate the future research in the acne healthcare community.

19 December 2023 in Knowledge-Based Systems
Verified
Journal Article
Closed access
XWXin Wei
Xin Wei
LZLei Zhang
Lei Zhang
JZJianwei Zhang
Jianwei Zhang
JWJunyou Wang
Junyou Wang
WLWenjie Liu
Wenjie Liu
XJXian Jiang
Xian Jiang

1 January 2011 in Biological and Pharmaceutical Bulletin
Verified
Journal Article
Open access
PDF file
GWGuangcheng Wang
Guangcheng Wang
BQBaowen Qi
Baowen Qi
HZHao Zheng
Hao Zheng
ZCZhizhi Chen
Zhizhi Chen
XWXin Wei
Xin Wei
LCLijuan Chen
Lijuan Chen
Idiopathic pulmonary fibrosis is regarded as a lethal chronic disease accompanied with excessive collagen disposition. In the early stage, monocyte chemotactic protein-1 (MCP-1) plays a crucial role in the process. Our previously screening with a vitro assay through inhibition of chemotaxis of RAW264.7 cells stimulated by MCP-1 proved that several analogues of thiazolidinediones, especially (Z)-5-(4-methoxybenzylidene)thiazolidine-2,4-dione (SKLB010), had potency of protecting acute liver injury in vivo without obvious toxicity. The present study aimed to investigate the preventive effect of SKLB010 in bleomycin-induced pulmonary fibrosis and further explore the underlying mechanisms. Bleomycin (BLM) was injected intratracheally at a single dose of 5 U kg−1 for pulmonary fibrosis induction. SKLB010 (25, 50 mg/kg/d) was respectively administrated by gavages 1 d prior to BLM administration and continued to the end of the study (for 4 weeks). Our results demonstrated that SKLB010 diminished the increase of macrophage, neutrophil and lymphocyte counts as well as the levels of tumor necrosis factor-α (TNF-α), interleukin (IL)-1β, and IL-6 in bronchoalveolar lavage fluid on day 14 (p<0.05). Moreover, oral gavages of SKLB010 also ameliorated histological changes and significantly suppressed collagen deposition on day 28. The treatment with SKLB010 exerted approximately 34.6% the hydroxyproline content reduction for 25 mg/kg dose and 56.7% reduction for 50 mg/kg dose in contrast to bleomycin-induced group (p<0.05). Meanwhile, SKLB010 inhibited the overexpression of tumor growth factor (TGF)-β1 and Smad3 in a dose-dependent manner. In conclusion, our results showed that SKLB010 could attenuate the BLM-induced pulmonary fibrosis in vivo and therefore be a promising anti-fibrogenic candidate.

10 June 2026 in Electronics
Verified
Journal Article
Open access
PDF file
ZTZhirong Tang
Zhirong Tang
XWXin Wei
Xin Wei
Zhaobin WeiZhaobin Wei
Zhaobin Wei
FTFei Tan
Fei Tan
CTCong Tian
Cong Tian
XXXuedou Xiong
Xuedou Xiong
Accurate estimation of the utility harmonic impedance at the Point of Common Coupling (PCC) is critical for harmonic pollution management in industrial power grids. Existing non-invasive methods rely heavily on restrictive assumptions that are rarely satisfied in practice, and conventional filtering-based approaches suffer from accuracy degradation in dynamic scenarios due to fixed-rule updates of the noise covariance. This paper proposes a deep reinforcement learning (RL)-optimized adaptive extended Kalman filter (AEKF) method for robust harmonic impedance estimation. A state-space model is established without restrictive assumptions, and a deep Q-network (DQN) framework is designed to optimize noise covariance updates adaptively. Simulation results show that the method achieves reliable estimation under normal conditions. Although errors rise under strong noise, it remains stable and exhibits better noise robustness than conventional methods. Field measurements in actual power grid environments further verified the feasibility and application potential of the proposed method in field engineering.

24 May 2022 in arXiv
Verified
Preprint
Open access
PDF file
LBLu Bai
Lu Bai
WJWeixing Ji
Weixing Ji
QLQinyuan Li
Qinyuan Li
XYXilai Yao
Xilai Yao
XWXin Wei
Xin Wei
WZWanyi Zhu
Wanyi Zhu
Deep learning is attracting interest across a variety of domains, including natural language processing, speech recognition, and computer vision. However, model training is time-consuming and requires huge computational resources. Existing works on the performance prediction of deep neural networks, which mostly focus on the training time prediction of a few models, rely on analytical models and result in high relative errors. %Optimizing task scheduling and reducing job failures in data centers are essential to improve resource utilization and reduce carbon emissions. This paper investigates the computational resource demands of 29 classical deep neural networks and builds accurate models for predicting computational costs. We first analyze the profiling results of typical networks and demonstrate that the computational resource demands of models with different inputs and hyperparameters are not obvious and intuitive. We then propose a lightweight prediction approach DNNAbacus with a novel network structural matrix for network representation. DNNAbacus can accurately predict both memory and time cost for PyTorch and TensorFlow models, which is also generalized to different hardware architectures and can have zero-shot capability for unseen networks. Our experimental results show that the mean relative error (MRE) is 0.9% with respect to time and 2.8% with respect to memory for 29 classic models, which is much lower than the state-of-the-art works.