A5101574191- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Endometrial and Cervical Cancer Treatments
- Radiomics and Machine Learning in Medical Imaging
- Autonomous Vehicle Technology and Safety
- Vehicle License Plate Recognition
- Meat and Animal Product Quality
- MRI in cancer diagnosis
- Energy Load and Power Forecasting
- Neural Networks and Applications
- Smart Grid Energy Management
- Cervical Cancer and HPV Research
- Advanced Image and Video Retrieval Techniques
- Microbial Inactivation Methods
- Machine Learning and ELM
- Image Retrieval and Classification Techniques
- Endometriosis Research and Treatment
- Magnetic and Electromagnetic Effects
- Paraquat toxicity studies and treatments
- Traditional Chinese Medicine Analysis
- Phytochemicals and Antioxidant Activities
- Food Supply Chain Traceability
- Stroke Rehabilitation and Recovery
- Internet Traffic Analysis and Secure E-voting
- Immune cells in cancer
Air Force General Hospital PLA
2024
Sichuan University
2023
Shandong Jiaotong University
2021-2023
Southern Medical University
2019-2022
Nanfang Hospital
2019-2022
Chongqing Maternal and Child Health Hospital
2019
Chinese Academy of Medical Sciences & Peking Union Medical College
2017-2018
Bohai University
2015-2016
In this study, a new predictive framework is proposed by integrating an improved grey wolf optimization (IGWO) and kernel extreme learning machine (KELM), termed as IGWO-KELM, for medical diagnosis. The IGWO feature selection approach used the purpose of finding optimal subset data. approach, genetic algorithm (GA) was firstly adopted to generate diversified initial positions, then (GWO) update current positions population in discrete searching space, thus getting better classification based...
Accurate multi-energy load forecasting for distributed systems is facing challenges due to the complexity of coupling and inherent stochasticity. In this regard, a novel stacking ensemble learning model based on reinforcement (RL) fusion proposed. First, feature selection performed using maximal information coefficient (MIC), data decomposed reconstructed through complete empirical mode decomposition with adaptive noise (CEEMDAN) sample entropy (SE). Subsequently, fused models strong...
Objective: To investigate whether pretreatment CT-derived radiomic features could be applied for prediction of clinical response to neoadjuvant chemotherapy (NACT) in locally advanced cervical cancer (LACC). Patients and Methods: 277 LACC patients treated with NACT followed by surgery/radiotherapy were included this multi-institution retrospective study. 1294 extracted from venous contrast enhanced non-enhanced CT imaging each patient. Five combined methods feature selection used reduce...
Abstract Purpose To develop and validate a deep learning‐based pathological risk score (RS) with an aim of predicting patients' prognosis to investigate the potential association between information within whole slide image (WSI) cervical cancer prognosis. Methods A total 251 patients International Federation Gynecology Obstetrics (FIGO) Stage IA1–IIA2 who underwent surgery without any preoperative treatment were enrolled in this study. Both clinical characteristics WSI each patient...
Vehicle Re-identification (Re-ID) refers to finding the same vehicle shot by other cameras from a given image library, which can also be regarded as sub-problem of retrieval. It plays an important role in intelligent transportation and smart cities. The key Re-ID is extract discriminative features. To better such features improve recognition accuracy, we propose three-branch adaptive attention network—Global Relational Attention Multi-granularity Feature Learning (GRMF) feature...
Effects of chitosan (CS) and hawthorn flavonoid coating on quality shelf life in flounder (Paralichthys olivaceus) fillets during refrigerated storage were investigated. Fish samples treated with 0.5% acetic aqueous, 1% CS, CS + 0.3% flavonoids, then stored at 4–8C for 14 days. All the analyzed periodically microbiological (total plate count psychrotrophic count), chemical (pH, TVB-N, K-value, TABRS), physical (hardness, color), sensory evaluation. Compared other two groups, flavonoids (CS...
Using thermal processing (TP) treatment (100 ℃, 1–8 min) as a control, the effects of high hydrostatic pressure (HHP, 200–500 MPa, 2.5–20 on microbiological and biochemical characteristics white shrimp Litopenaeus vannamei were investigated. The results showed that efficiency polyphenol oxidase (PPO) inactivation log reduction total plate count (TPC) by HHP all significantly lower than TP ( p < 0.05). rate for TPC PPO increased with increase holding time was in accordance first-order...
Agricultural food's logistics needs to be efficient and provide assurance on the safety quality of its products which consumers could trust. This paper designs a machine vision system by fruits or vegetables can detected for defects damages during transportation storage. The color histogram extracted in local image patch is used as feature Linear SVM (Support vector machine) model learning, provides good robustness, higher accuracy modest calculation costs. In case apple inspection, our...
Background: We aimed to investigate whether pre-therapeutic radiomic features based on magnetic resonance imaging (MRI) can predict the clinical response neoadjuvant chemotherapy (NACT) in patients with locally advanced cervical cancer (LACC).Methods: A total of 275 LACC receiving NACT were enrolled this study from eight hospitals, and allocated primary independent validation cohorts (2:1 ratio). Three feature sets extracted intratumoural region T1-weighted images, T2-weighted peritumoural...
ABSTRACT Objective: The objective of this study was to analyze the current status barriers exercise participation (EP) among patients on maintenance hemodialysis (MHD). Materials and Methods: A cross-sectional conducted 277 outpatients undergoing MHD in 2 tertiary first-class hospitals Beijing from February 2023 June who were selected using convenience sampling method. data collected general information questionnaire, Physical Activity Rating Scale, Exercise Benefits/Barriers Scale (EBBS),...
Ensuring the network security, resists malicious traffic attacks as much possible, and ensuring Gated Recurrent Unit (GRU) Convolutional Neural Network (CNN) are combined. Then, a Software Defined Networking (SDN) anomaly detection architecture is built continuously optimized to ensure security possible enhance reliability of architecture. The results show that proposed can greatly improve accuracy detection, its performance will be different due number CNN layers. When two-layer structure...
Objective: To explore a novel approach for the isolation of human endometriotic stem cells further study in cell theory endometriosis. Materials and Methods: A serum-free medium (SFM) was used to isolate from ectopic endometrium, eutopic normal endometrium vitro, biological properties including clonogenicity, multipotency, invasive ability were examined compared. Results: The endometrium-derived cultured SFM showed higher cloneforming efficiency sphere formation (SFE) than they all can...
An algorithm that can effectively distinguish different vehicles with high similarity by identifying vehicle photos collected from multiple angles is called recognition algorithm. This has been extensively applicable to the region of intelligent transportation and urban computing, but it always challenging implement In this paper, an modified feature extraction method provided, which enhances traditional nonlocal neural networks improves its ability capture relationship between positions in...