Xiujuan Zheng

Wuhan University of Science and Technology

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Kai Liu

Nanjing Audit University

K.Y. Tong

Chinese University of Hong Kong

Rong Song

Sun Yat-sen University

Xiaoling Hu

Hong Kong Polytechnic University

Chengxiang Xia

Institute for Stem Cell Biology and Regenerative Medicine

Dagan Feng

University of Sydney

Dehao Huang

Southern University of Science and Technology

Qitong Weng

Chinese Academy of Sciences

Fangxiao Hu

Institute for Stem Cell Biology and Regenerative Medicine

Yunqing Lin

Chinese Academy of Sciences

Cooperative Institutions

Chinese Academy of Sciences

Sichuan University

Shanghai Jiao Tong University

University of Chinese Academy of Sciences

Institute of Zoology

Fudan University

Guangzhou Institutes of Biomedicine and Health

Hong Kong Polytechnic University

Wuhan University of Science and Technology

Zhejiang University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Three-dimension deep model for body mass index estimation from facial image sequences with different poses

Journal of Visual Communication and Image Representation (2024)

Xiang Cui Boxiang Liu Liang Zhao Xiujuan Zheng

10.1016/j.jvcir.2024.104381

Cite

Citations (0)

A Novel Virtual Reality Design of Portable Automatic Perimetry

2022 IEEE MTT-S International Microwave Biomedical Conference (IMBioC) (2019)

Jia He Sheng Zhang Peixuan Wu Yun Zhang Xiujuan Zheng

Due to delayed diagnosis and treatment, millions of people, especially senior citizens, suffer from visual field loss even long-term blindness. However, many blindness could be prevented by popularizing the routine inspection of visual field. In this paper, we propose a novel virtual reality design of portable automatic perimetry. The virtual reality (VR) device provides a non-light and full-field environment. Combined with its portability, it makes external light shielded, and full-field examination feasible in normal clinic ward or outpatient department as a kind of regular physical examination. To make it easier to use, we also offered several automatic human-VR interaction modes, including using straightforward eye tracking to replace traditional mouse click or button press. Owing to more friendly interaction and better portability, the novel virtual reality design of portable automatic perimetry facilitates both the subjects and clinicians to improve the efficiency of routine inspection.

Software portability

Optical head-mounted display

10.1109/imbioc.2019.8777783

Cite

Citations (3)

[An automatic 3D brain segmentation based on improved level-set method in micro-CT rat/mouse images].

PubMed (2012)

Shiye Chen Cheng Wang Xiujuan Zheng

In vivo Micro-PETICT imaging of mouse/rat brain has been widely used to non-invasively monitor brain and provides researchers a better understanding of therapeutic effects in models of human neurological disease. For the need of further processing, extraction of brain tissue from head is required and vital.An automatic multistep combination methods was proposed based on an improved level set framework, which includes (1) Use Fuzzy-C-Means method together with threshold and morphology methods to get the initial level-set surface automatically. (2) Combine gradient vector flow to enhance the gradient contrast and enforce the surface move toward to the object's surface much faster, especially obtain a significantly improvement in the regions of forehead and the joint between brain and neck. (3) introduce an automatic stop condition based on average bandwidth energy maximization to overcome the leakage problem.3 Micro-CT images of rat and 3 of mouse have been tested using the proposed methods and the average accuracy has increased by 33% for rat and 6.7% for mouse. The average processing duration for rat and mouse are about 8 minutes and 4 minutes, respectively.The proposed methods were proved that it can be effectively used for Micro-PET/CT imaging of mouse/rat brain segmentation and have a great improvement on accuracy and convenience.

Source

Cite

Citations (0)

Engineered CRO-CD7 CAR-NK cells derived from pluripotent stem cells avoid fratricide and efficiently suppress human T-cell malignancies

bioRxiv (Cold Spring Harbor Laboratory) (2025)

Yunqing Lin Zi-Yun Xiao Fangxiao Hu Xiujuan Zheng Chenyuan Zhang

Background: T cell malignancies are highly aggressive hematological tumors with limited effective treatment options. CAR-NK cell therapy targeting CD7 has emerged as a promising approach for treating T-cell malignancies. However, conventional CAR-NK cell therapy faces the challenges of cell fratricide due to CD7 expression on both malignant cells and normal NK cells. Additionally, engineering CARs into human tissue-derived NK cells demonstrates heterogeneity, low transduction efficiency, and high manufacturing costs. Methods: The human pluripotent stem cells (hPSCs) were genetically modified by knocking out the CD7 gene and introducing the CD7 CAR expression cassette to generate CD7 KO-CD7 CAR-hPSCs. These modified hPSCs were subsequently differentiated into CD7 KO-CD7 CAR-iNK cells using an efficient organoid induction method. The cytotoxicity of CD7 KO-CD7 CAR-iNK cells against CD7+ tumor cells was evaluated. Furthermore, we overexpressed the CXCR4 gene in CD7 KO-CD7 CAR-hPSCs and derived CXCR4-expressing CD7 KO-CD7 CAR-iNK (CRO-CD7 CAR-iNK) cells. The dynamics of CRO-CD7 CAR-iNK cells in vivo were tracked, and their therapeutic efficacy was assessed using human T-cell acute lymphoblastic leukemia (T-ALL) xenograft models. Results: The CD7 KO-CD7 CAR-iNK cells derived from CD7 KO-CD7 CAR-hPSCs effectively avoided fratricide, demonstrated normal expansion, and exhibited potent and specific anti-tumor activity against CD7+ T-cell tumor cell lines and primary T-ALL cells. CXCR4 overexpression in CRO-CD7 CAR-iNK cells improved their homing capacity and extended their persistence in vivo. The CRO-CD7 CAR iNK cells significantly suppressed tumor growth and prolonged the survival of T-ALL tumor-bearing mice. Conclusions: Our study provides a reliable strategy for the large-scale generation of fratricide-resistant CD7 CAR-iNK cells with robust anti-tumor effects from hPSCs, offering a promising cell product to treat T-cell malignancy.

10.1101/2025.01.23.634617

Cite

Citations (0)

Multirate Signal based Fault Diagnosis of Hydraulic Systems with Residual Shrinkage Separable Convolutional Network

2021 China Automation Congress (CAC) (2024)

Pengpeng Qu Yong Zhang Rui Tang Yong Ge Xiujuan Zheng

Shrinkage

SIGNAL (programming language)

10.1109/cac63892.2024.10864759

Cite

Citations (0)

ADCI-Net: an adaptive discriminative clip identification strategy for fast video action recognition

Journal of Electronic Imaging (2021)

Zizhao Guo Xiujuan Zheng Xinyi Chun Sancong Ying

The most common method for video-level classification is to operate models over a whole series of fixed-temporal-length clips in a long video. However, a video usually consists of several meaningless sections, and similar clips that are fed into models could lead to a suboptimal recognition accuracy and waste of computing resources. To address this issue, we introduced an adaptive discriminative clip identification network to evaluate every video clip with respect to its relevance. We adaptively choose top-ranked clips as inputs for prediction and filter out irrelevant clips. Specifically, for a given trained cumbersome convolutional neural network action recognition model, we use a lightweight hallucination network (H-Net) to study its generalization ability based on distillation. Then, the evaluation of relevant clips is performed by considering the imitated features of H-Net. Thus, heavy calculations of overlapping content or meaningless clips can be avoided during inferences. We validate our approach by examining its performance on two datasets: UCF101 and HMDB51. Our strategy can be applied to any clip-based action recognition model. Experimental results demonstrate that on UCF101, we reduced the computational cost by 70% and increased the accuracy.

Discriminative model

CLIPS

Identification

10.1117/1.jei.30.2.023030

Cite

Citations (2)

Study on Random Decrement Signature Under Different Triggering Level and Length of Time History

Conference proceedings of the Society for Experimental Mechanics (2017)

Jinzhi Wu Xiujuan Zheng Jie Hu Yigang Zhang Fuh-Gwo Yuan

Signature (topology)

Random vibration

Identification

10.1007/978-3-319-54810-4_17

Cite

Citations (0)

A Comparison Between Electromyography-Driven Robot and Passive Motion Device on Wrist Rehabilitation for Chronic Stroke

Neurorehabilitation and neural repair (2009)

Xiaoling Hu K.Y. Tong Rong Song Xiujuan Zheng Wallace Woon‐Fong Leung

Background. The effect of using robots to improve motor recovery has received increased attention, even though the most effective protocol remains a topic of study. Objective . The objective was to compare the training effects of treatments on the wrist joint of subjects with chronic stroke with an interactive rehabilitation robot and a robot with continuous passive motion. Methods. This study was a single-blinded randomized controlled trial with a 3-month follow-up. Twenty-seven hemiplegic subjects with chronic stroke were randomly assigned to receive 20-session wrist training with a continuous electromyography (EMG)-driven robot (interactive group, n = 15) and a passive motion device (passive group, n = 12), completed within 7 consecutive weeks. Training effects were evaluated with clinical scores by pretraining and posttraining tests (Fugl-Meyer Assessment [FMA] and Modified Ashworth Score [MAS]) and with session-by-session EMG parameters (EMG activation level and co-contraction index). Results. Significant improvements in FMA scores (shoulder/elbow and wrist/hand) were found in the interactive group ( P < .05). Significant decreases in the MAS were observed in the wrist and elbow joints for the interactive group and in the wrist joint for the passive group ( P < .05). These MAS changes were associated with the decrease in EMG activation level of the flexor carpi radialis and the biceps brachii for the interactive group ( P < .05). The muscle coordination on wrist and elbow joints was improved in the interactive groups in the EMG co-contraction indexes across the training sessions ( P < .05). Conclusions. The interactive treatment improved muscle coordination and reduced spasticity after the training for both the wrist and elbow joints, which persisted for 3 months. The passive mode training mainly reduced the spasticity in the wrist flexor.

Stroke

10.1177/1545968309338191

Cite

Citations (196)

Cerebral blood perfusion changes in amputees with myoelectric hands after rehabilitation: a SPECT computer-aided analysis

BMC Neuroscience (2016)

Qiufang Liu Xiujuan Zheng Panli Li Lian Xu Longwen He

Rehabilitation, which is essential for amputees with myoelectric hands, can improve the quality of daily life by remodeling the neuron network. In our study, we aim to develop a cerebral blood perfusion (CBF) single-photon emission computed tomography computer-aided (SPECT-CA) detection scheme to automatically locate the brain's activated regions after rehabilitation. Five participants without forearms (three male, two female, mean age 51 ± 12.89 years, two missing the right side, and three missing the left side) were included in our study. In the clinical assessment, all of the participants received higher scores after training. The results of the SPM analysis indicated that CBF in the precentral gyrus, postcentral gyrus, frontal lobe, temporal lobe and cerebellum was significantly different among the five participants (P < 0.05). Moreover, SPECT-CA showed that the activated brain areas mainly included the precentral gyrus, postcentral gyrus, cerebellum and extensive cerebral cortex. Our study demonstrated that the CBF SPECT-CA method can detect the brain blood perfusion changes induced by rehabilitation with high sensitivity and accuracy. This method has great potential for locating the remodeled neuron regions of amputees with myoelectric hands after rehabilitation.

Postcentral gyrus

Precentral gyrus

Frontal lobe

Supplementary motor area

Middle frontal gyrus

Parietal lobe

10.1186/s12868-016-0294-3

Cite

Citations (4)

Scale-Adaptive Deep Model for Bacterial Raman Spectra Identification

IEEE Journal of Biomedical and Health Informatics (2021)

Lin Deng Yuzhong Zhong Maoning Wang Xiujuan Zheng Jianwei Zhang

The combination of Raman spectroscopy and deep learning technology provides an automatic, rapid, and accurate scheme for the clinical diagnosis of pathogenic bacteria. However, the accuracy of existing deep learning methods is still limited because of the single and fixed scales of deep neural networks. We propose a deep neural network that can learn multi-scale features of Raman spectra by using the automatic combination of multi-receptive fields of convolutional layers. This model is based on the expert knowledge that the discrimination information of Raman spectra is composed of multi-scale spectral peaks. We enhance the interpretability of the model by visualizing the activated wavenumbers of the bacterial spectrum that can be used for reference in related work. Compared with existing state-of-the-art methods, the proposed method achieves higher accuracy and efficiency for bacterial identification on isolate-level, empiric-treatment-level, and antibiotic-resistance-level tasks. The clinical bacterial identification task requires significantly fewer patient samples to achieve similar accuracy. Therefore, this method has tremendous potential for the identification of clinical pathogenic bacteria, antibiotic susceptibility testing, and prescription guidance.

Interpretability

Identification

10.1109/jbhi.2021.3113700

Cite

Citations (28)