Zixuan Cheng

ORCID: 0000-0001-6761-8583
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Computing and Algorithms
  • Brain Tumor Detection and Classification
  • Radiomics and Machine Learning in Medical Imaging
  • Web Data Mining and Analysis
  • Dental Radiography and Imaging
  • Advanced Text Analysis Techniques
  • Computational Drug Discovery Methods
  • Text and Document Classification Technologies
  • Orthodontics and Dentofacial Orthopedics
  • Anomaly Detection Techniques and Applications
  • Forensic Anthropology and Bioarchaeology Studies
  • Melanoma and MAPK Pathways
  • Soil Mechanics and Vehicle Dynamics
  • Microbial Natural Products and Biosynthesis
  • Educational Technology and Pedagogy
  • Industrial Technology and Control Systems
  • Data Stream Mining Techniques
  • Advanced Neural Network Applications
  • Tribology and Wear Analysis
  • Embedded Systems and FPGA Design
  • Non-Invasive Vital Sign Monitoring
  • Parallel Computing and Optimization Techniques
  • Bioinformatics and Genomic Networks
  • Cloud Computing and Remote Desktop Technologies
  • Emotion and Mood Recognition

Changchun University
2014-2024

Swinburne University of Technology Sarawak Campus
2023

Stomatological Hospital of Chongqing Medical University
2023

Chongqing Medical University
2023

Jilin University
2020-2021

Guangzhou First People's Hospital
2020

Guangzhou Medical University
2020

Guangdong Provincial People's Hospital
2019

South China University of Technology
2018-2019

University of Science and Technology Beijing
2018

To evaluate the techniques used for automatic digitization of cephalograms using artificial intelligence algorithms, highlighting strengths and weaknesses each one reviewing percentage success in localizing cephalometric point.Lateral were digitized traced by three calibrated senior orthodontic residents with or without (AI) assistance. The same radiographs 43 patients uploaded to AI-based machine learning programs MyOrthoX, Angelalign, Digident. Image J was extract x- y-coordinates 32...

10.1186/s12903-023-03188-4 article EN cc-by BMC Oral Health 2023-07-08

Methylation of the O 6 -methylguanine methyltransferase (MGMT) gene promoter is correlated with effectiveness current standard care in glioblastoma patients. In this study, a deep learning pipeline designed for automatic prediction MGMT status 87 patients contrast-enhanced T1W images and 66 fluid-attenuated inversion recovery(FLAIR) images. The end-to-end completes both tumor segmentation classification. better performance comes from FLAIR (Dice score, <mml:math...

10.1155/2020/9258649 article EN cc-by BioMed Research International 2020-09-23

Protein kinase p38γ is an attractive target against cancer because it plays a pivotal role in cell proliferation by phosphorylating the retinoblastoma tumour suppressor protein. Therefore, inhibition of with active small molecules represents alternative for developing anti-cancer drugs. In this work, we present rigorous and systematic virtual screening framework to identify potential inhibitors cancer. We combined use machine learning-based quantitative structure activity relationship...

10.3390/ijms24087360 article EN International Journal of Molecular Sciences 2023-04-17

Public safety issues have always been the focus of widespread concern people from all walks life. With development video detection technology, abnormal human behavior in videos has become key to preventing public issues. Particularly, student groups, is very important. Most existing algorithms are aimed at outdoor activity detection, and indoor effects these not ideal. Students spend most their time indoors, modern classrooms mostly equipped with monitoring equipment. This study focuses on...

10.1155/2020/6625695 article EN Mathematical Problems in Engineering 2020-11-30

Purpose The purpose of this paper is to synthesize a microemulsion and investigate its tribological properties as lubricant. Magnesium alloy warm rolling experiments were conducted. Surface morphology was observed wear form summarized. composition surface residues analyzed, which sheds light on the lubrication mechanism microemulsion. Design/methodology/approach A prepared with proper amount oil, surfactant, cosurfactant, water other additives for magnesium strip rolling. Tribological...

10.1108/ilt-06-2018-0222 article EN Industrial Lubrication and Tribology 2018-11-26

With the rapid development of we-media information dissemination, WeChat official accounts platform has become an important way for people to obtain health related knowledge. However, is redundant, miscellaneous, and overloaded. In order meet increasingly accurate efficient knowledge service needs users, reorganizing aggregating document resources effective. If we use artificial recognition filter information, it will inevitably cause huge labor time cost, effect very little in front massive...

10.1155/2022/1166989 article EN cc-by Journal of Environmental and Public Health 2022-01-01

Based on the characteristics of convenience, autonomy, and equality, online self-media has become an important way for contemporary migrant workers to observe world, understand society, examine themselves express their demands. On basis analysis domestic works' concerns emotion analysis, we crawl data Weibo about topics as basic corpus concerns, then uses a combination TF-IDF Word2Vec methods construct recognition model workers' concerns. We found that wages, children's education, medical...

10.3389/fpsyg.2021.741928 article EN cc-by Frontiers in Psychology 2021-08-31

To investigate the effects of different wavelet filters on correlation and diagnostic performance radiomics features. Methods: A total 143 colorectal cancer (CRC) patients (64 positive in lymph node metastasis 79 negative) with contrast-enhanced CT examination were recruited. After labeling tumor area by experienced radiologists, wavelets features based 48 extracted using in-house software coded Matlab. The coefficients same names between calculated got distribution high-correlation each...

10.11817/j.issn.1672-7347.2019.03.003 article EN PubMed 2019-03-28

Depression is a prevalent mental condition, and it essential to diagnose treat patients as soon possible maximize their chances of rehabilitation recovery. An intelligent detection model based on multimodal fusion technology proposed the findings this study address difficulties associated with depression detection. Text data electroencephalogram (EEG) are used in representatives subjective objective nature, respectively. These processed by BERT–TextCNN CNN–LSTM model, which responsible for...

10.1142/s0219519424400463 article EN Journal of Mechanics in Medicine and Biology 2024-07-18

Mental health is critical to an individual’s life and social functioning affects emotions, cognition behavior. status assessments can help individuals understand their own psychological status, identify potential problems in real-time implement effective interventions promote favorable mental health. In this study, a deep learning approach was used construct simple-minded flexible model for electroencephalogram (EEG)-based assessment the corresponding M4EEG model. This suitable not only...

10.1142/s0219519424400372 article EN Journal of Mechanics in Medicine and Biology 2024-07-18

Polo-like kinase 1 (PLK1) and p38γ mitogen-activated protein (p38γ) play important roles in cancer pathogenesis by controlling cell cycle progression are therefore attractive targets. The design of multitarget inhibitors may offer synergistic inhibition distinct targets reduce the risk drug–drug interactions to improve balance between therapeutic efficacy safety. We combined deep-learning-based quantitative structure–activity relationship (QSAR) modeling hybrid-based consensus scoring screen...

10.1021/acs.jcim.3c01252 article EN Journal of Chemical Information and Modeling 2023-10-26

Time series data are an extremely important type of in the real world. gradually accumulate over time. Due to dynamic growth time data, they tend have higher dimensions and large scales. When performing cluster analysis on this there shortcomings using traditional feature extraction methods for processing. To improve clustering performance study uses a recurrent neural network (RNN) train input data. First, RNN called long short-term memory (LSTM) is used extract features Second, pooling...

10.1155/2020/8885625 article EN Mathematical Problems in Engineering 2020-10-13

Advances in medical imaging skills have promoted the influence of neuroscience. Having advanced technology is essential for industry. Magnetic resonance (MRI) plays a central role imaging. It key treatment various human diseases. Doctors analyze brain size, shape, and location MR images to assess disease develop plan. The manual division tissue by experts heavy subjective. Therefore, study automatic segmentation has practical significance. Because characteristics MRI are low contrast high...

10.1166/jmihi.2021.3285 article EN Journal of Medical Imaging and Health Informatics 2020-10-28

With the popularization and application of computer, computer desk is much higher.In order to meet requirements new features, its structure also constantly changing.In this paper, deficiencies current on market have been focused analysis.Structural design has begun, comfortableness designed, in make it ergonomics.After desk, was compact, shaping beautiful, can user convenient comfortable.

10.2174/1874110x01408010998 article EN The Open Cybernetics & Systemics Journal 2014-12-31

10.1007/s12652-021-03125-9 article EN Journal of Ambient Intelligence and Humanized Computing 2021-03-24

Based on existing crop production positioner analytical investigation and according to working principle of positioner, this study did a research transmission design. The paper mainly designed worm gear retarding mechanism table rotary direction. Aiming at mechanical structure related parts were calculated checked, including checking calculation keyed shaft, selection sliding bearing. Calculated results are certain theoretical significance for design study.

10.19026/ajfst.11.2366 article EN Advance Journal of Food Science and Technology 2016-04-05

Abstract Background : To evaluate the techniques used for automatic digitization of cephalograms, highlighting strengths and weaknesses each one review percentage success in localising cephalometric point.、 Methods Lateral cephalograms were digitized traced by three calibrated senior orthodontic residents with or without artificial intelligence (AI) assistance. The same radiographs 43 patients uploaded to AI-based machine learning program MyOrthoX, Angelalign Digident. Image J was extract x-...

10.21203/rs.3.rs-2508746/v1 preprint EN cc-by Research Square (Research Square) 2023-02-07

With the advent of era big data, number various knowledge resources has exploded. A series problems, such as information redundancy and overload, bring trouble to users ' reading use. The generation text summary can improve efficiency user acquisition reuse rate, realize integration orderly organization. This paper integrates automatic summarization technology into service. Firstly, a method based on improved TextRank algorithm is proposed. Then, service elements are defined. Finally, model...

10.1145/3644523.3644591 article EN 2023-10-13

Because the FCM method is simple and effective, a series of research results based on this are widely used in medical image segmentation. Compared with traditional FCM, probability clustering (PCM) algorithm cancels constraint normalization each sample membership degree iterative process, effect improved within certain range. However, above two methods only use gray value pixels ignoring context relationship between high-dimensional pixels. The easily affected by noise during segmentation...

10.1166/jmihi.2020.3093 article EN Journal of Medical Imaging and Health Informatics 2020-04-17

Instruction set simulator is a critical tool for both architecture exploration and design of software. With the development embedded system, more complex software platform required. An instruction specific to micro-controller with different shall be developed. In this paper, approaches implement an ISS have been divided into traditional advanced approaches. About five discussed compared use cases. end, interpretive simulation technique suggested its full flexibility popularity in practice.

10.1109/eiecs53707.2021.9588034 article EN 2021 International Conference on Electronic Information Engineering and Computer Science (EIECS) 2021-09-23
Coming Soon ...