Cong Wei

ORCID: 0000-0003-4660-6916
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Renal cell carcinoma treatment
  • Cancer Genomics and Diagnostics
  • Domain Adaptation and Few-Shot Learning
  • Multimodal Machine Learning Applications
  • Insect-Plant Interactions and Control
  • Thyroid Cancer Diagnosis and Treatment
  • Topic Modeling
  • Hydrocarbon exploration and reservoir analysis
  • Head and Neck Anomalies
  • Metabolomics and Mass Spectrometry Studies
  • Hydraulic Fracturing and Reservoir Analysis
  • Privacy-Preserving Technologies in Data
  • Neurological Disease Mechanisms and Treatments
  • COVID-19 diagnosis using AI
  • Ophthalmology and Visual Impairment Studies
  • Cancer-related molecular mechanisms research
  • CAR-T cell therapy research
  • Connective Tissue Growth Factor Research
  • Total Knee Arthroplasty Outcomes
  • Shoulder and Clavicle Injuries
  • Acne and Rosacea Treatments and Effects
  • Autophagy in Disease and Therapy
  • Amyotrophic Lateral Sclerosis Research
  • Traditional Chinese Medicine Analysis
  • Thyroid and Parathyroid Surgery

Tarim University
2017-2024

Guangzhou Medical University
2024

China National Petroleum Corporation (China)
2024

Sun Yat-sen University
2024

Research Institute of Petroleum Exploration and Development
2024

China Resources (China)
2024

South China University of Technology
2024

Shanghai University
2023

Dalian Medical University
2023

State Key Laboratory of Robotics
2023

Federated learning-based semantic segmentation (FSS) has drawn widespread attention via decentralized training on local clients. However, most FSS models assume categories are fixed in advance, thus heavily undergoing forgetting old practical applications where clients receive new incrementally while have no memory storage to access classes. Moreover, collecting novel classes may join the global of FSS, which further exacerbates catastrophic forgetting. To surmount above challenges, we...

10.1109/cvpr52729.2023.00383 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023-06-01

Stroke, especially ischemic stroke, is an important cause of neurological morbidity and mortality worldwide. Growing evidence suggests that the immune system plays intricate function in pathophysiology stroke. Gelsevirine (Gs), alkaloid from Gelsemium elegans, has been proven to decrease inflammation neuralgia osteoarthritis previously, but its role stroke unknown. In this study, middle cerebral artery occlusion (MCAO) mice model was used evaluate protective effect Gs on administration...

10.3389/fimmu.2023.1164278 article EN cc-by Frontiers in Immunology 2023-03-30

Previous studies have shown that children with intermittent exotropia (IXT) a higher rate of psychiatric abnormalities as they grow up, such attention deficit. This study explored visual and hearing among IXT, evaluated its association clinical characteristics cognitive development. Forty-nine diagnosis IXT 29 traditional development were recruited. The Integrated Visual Auditory Continuous Performance Test (IVA-CPT) was used to measure the subjects' full-scale response control quotient...

10.1186/s13052-024-01591-3 article EN cc-by ˜The œItalian Journal of Pediatrics/Italian journal of pediatrics 2024-01-25

Toxic gain-of-function mutations in superoxide dismutase 1 (SOD1) contribute to approximately 2%-3% of all amyotrophic lateral sclerosis (ALS) cases. Artificial microRNAs (amiRs) delivered by adeno-associated virus (AAV) have been proposed as a potential treatment option silence SOD1 expression and mitigate disease progression. Primary microRNA (pri-miRNA) scaffolds are used amiRs shuttle hairpin RNA into the endogenous miRNA pathway, but it is unclear whether different primary impact...

10.1016/j.omtn.2023.102057 article EN cc-by-nc-nd Molecular Therapy — Nucleic Acids 2023-10-16

Continual Named Entity Recognition (CNER) is a burgeoning area, which involves updating an existing model by incorporating new entity types sequentially. Nevertheless, continual learning approaches are often severely afflicted catastrophic forgetting. This issue intensified in CNER due to the consolidation of old from previous steps into non-entity type at each step, leading what known as semantic shift problem type. In this paper, we introduce pooled feature distillation loss that...

10.18653/v1/2023.emnlp-main.509 article EN cc-by Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing 2023-01-01

Using the ensemble learning method to mine valuable information from a sea of financial data accumulated on market securities is very important for studying processing. On basis A-share companies listed Shanghai Stock Market, this article takes perspective unbalanced classification ST stocks carry out study construction warning model companies. In our experiment, HDRF (HDRandom Forest, Hellinger Distance based Random Forest), models Bagging, AdaBoost, and Rotation which take distance...

10.4018/ijiit.2020010103 article EN International Journal of Intelligent Information Technologies 2019-11-25

Objective To develop an easy‐to‐use nomogram for discrimination of malignant thyroid nodules and to compare diagnostic efficiency with the Kwak American College Radiology (ACR) Thyroid Imaging, Reporting Data System (TI‐RADS). Study Design Retrospective study. Setting The Second Hospital Shandong University. Subjects Methods From March 2017 April 2019, 792 patients 1940 were included into training set; from May 2019 December 174 389 validation set. Multivariable logistic regression model was...

10.1177/0194599820939071 article EN Otolaryngology 2020-07-21

The aim of the present study was to investigate whether ultrasound treatment combined with microbubbles inhibits cell invasion and migration in androgen‑independent prostate cancer (PCa) cells identify probable mechanism. Ultrasound used continuous wave mode at a frequency 21 kHz spatial‑average temporal‑average intensity 46 mW/cm2. (200 µl; SonoVue) administered human PCa PC‑3 for 30 sec. were divided into three groups: control group, group (US) (US + MB). Following 12, 24, 48 72 h,...

10.3892/ol.2014.2310 article EN Oncology Letters 2014-07-03

The present study tries to identify proteins implicated in bradycardia rabbits hearts after ShenSongYangXin (SSYX, a traditional Chinese medicine) treatment. Eighteen adult were randomly assigned three groups: sham, model, and SSYX treatment groups. Heart rate was recorded isolated from ventricular muscle. We used isobaric tags for elative absolute quantitation (iTRAQ) coupled with two-dimensional liquid chromatography-tandem mass spectrometry altered heart decreased six weeks due the injury...

10.1155/2015/385953 article EN Evidence-based Complementary and Alternative Medicine 2015-01-01

Papillary thyroid carcinoma (PTC) is one of the fastest growing endocrine system malignant carcinomas detected over past decade.Unfortunately, more than 25% PTC patients are characterized by their aggressiveness and subsequent metastasis; these characteristics usually indicate poor prognosis.Recently, increasing evidence has suggested that solute carrier (SLC) transporters may play a pivotal role in initiation, invasion metastasis human carcinoma.However, expression clinicopathological...

10.7150/jca.56730 article EN cc-by-nc Journal of Cancer 2021-01-01

Aims. The present study tries to investigate the gene expression profile of bradycardia rabbits' hearts after SSYX (SSYX, a traditional Chinese medicine) treatment. Methods. Eighteen adult rabbits were randomly assigned in three groups: sham, model, and treatment groups. Heart rate was recorded total RNA isolated from hearts. Gene profiling conducted quantitative real-time reverse transcription-polymerase chain reaction (RT-PCR) performed confirm results. Patch clamp using human induced...

10.1155/2014/715937 article EN cc-by Evidence-based Complementary and Alternative Medicine 2014-01-01

Shensongyangxin (SSYX) is a medicinal herb, which has long been used in traditional Chinese medicine. Various pharmacological activities of SSYX have identified. However, the role cardiac hypertrophy remains to be fully elucidated. In present study, aortic banding (AB) was performed induce mice. (520 mg/kg) administered by daily gavage between 1 and 8 weeks following surgery. The extent then evaluated pathological molecular analyses heart tissue samples. addition, vitro experiments were...

10.3892/mmr.2015.4598 article EN Molecular Medicine Reports 2015-11-23

The accurate, quantitative, and objective prediction of the brain age for premature infants will contribute to exploration maturity catch-up growth. Traditional approaches rely heavily on a pediatrician’s clinical experience, which makes whole process time-consuming labor-intensive. To solve this problem, we propose deep learning-based model preterm via neonatal MRI purpose, it is called as BAPNET short. First all, collected specific dataset including MR images 281 infants. Then, pretraining...

10.1109/access.2023.3291810 article EN cc-by IEEE Access 2023-01-01

Abstract Family with sequence similarity 20‐member C (FAM20C) is a kinase specific to most of the secreted phosphoproteome. FAM20C has been identified as causative gene Raine syndrome, initially characterized by lethal osteosclerosis bone dysplasia. However, since identification cases nonlethal syndrome hypophosphatemia rickets, previous definition become debatable and raised question about role mutations in controversial skeletal manifestation two forms disease. In this study, we aimed...

10.1002/jcp.31105 article EN Journal of Cellular Physiology 2023-09-12

Objective: Minimally invasive video-assisted technique has been successfully used in thyroidectomy to treat low-and intermediate-risk papillary thyroid cancer.But there was very little experience of this be the lateral neck lymphadenectomy cancer with lymph node metastasis.This study aimed verify potential benefits a new minimally that evaluated likelihood surgical method as an alternative for carcinoma metastasis.Method: A total 58 patients were retrospectively compared, which included two...

10.1093/jjco/hyw055 article EN Japanese Journal of Clinical Oncology 2016-05-10

Incremental semantic segmentation endeavors to segment newly encountered classes while maintaining knowledge of old classes. However, existing methods either 1) lack guidance from class-specific (i.e., class prototypes), leading a bias towards new classes, or 2) constrain class-shared model weights) excessively without discrimination, resulting in preference for In this paper, trade off performance, we propose the Class-specific and Class-shared Knowledge (Cs2K) incremental segmentation....

10.48550/arxiv.2407.09047 preprint EN arXiv (Cornell University) 2024-07-12
Coming Soon ...