A5100393862- Scheduling and Optimization Algorithms
- Cancer Immunotherapy and Biomarkers
- Advanced Image Processing Techniques
- Advanced Manufacturing and Logistics Optimization
- Lung Cancer Treatments and Mutations
- Power Systems and Renewable Energy
- Image Enhancement Techniques
- Medical Imaging and Analysis
- Industrial Automation and Control Systems
- Photoacoustic and Ultrasonic Imaging
- AI in cancer detection
- Antenna Design and Optimization
- Colorectal Cancer Treatments and Studies
- Speech and Audio Processing
- Traffic control and management
- Graph Theory and Algorithms
- Advanced Image Fusion Techniques
- Civil and Geotechnical Engineering Research
- COVID-19 diagnosis using AI
- Millimeter-Wave Propagation and Modeling
- Cancer Research and Treatments
- Metastasis and carcinoma case studies
- Traffic Prediction and Management Techniques
- Distributed Sensor Networks and Detection Algorithms
- Advanced Graph Neural Networks
China Construction Bank
2024
Jiaxing University
2023
Guangdong Medical College
2021-2022
Affiliated Hospital of Guizhou Medical University
2022
Guiyang Medical University
2022
Affiliated Hospital of Guangdong Medical College Hospital
2019
IntroductionSintilimab plus chemotherapy significantly prolonged progression-free survival (PFS) compared with alone in nonsquamous NSCLC the ORIENT-11 study. Updated overall (OS) and PFS data corresponding biomarker analyses are reported here.MethodsIn this study, a total of 397 patients previously untreated, locally advanced or metastatic were assigned to sintilimab combination treatment (combo) group placebo group. The stratified by programmed death-ligand 1 (PD-L1) expression levels....
Machine learning has begun to play a central role in many applications. A multitude of these applications typically also involve datasets that are distributed across multiple computing devices/machines due either design constraints (e.g., multiagent systems) or computational/privacy reasons on smartphone data). Such often require the tasks be carried out decentralized fashion, which there is no server directly connected all nodes. In real-world settings, nodes prone undetected failures...
A hybrid flow shop scheduling model with missing and re-entrant operations was designed to minimize the maximum completion time reduction in energy consumption. The proposed dual-population genetic algorithm enhanced a range of improvements, which include design three-layer gene coding method, hierarchical crossover mutation techniques, development an adaptive operator that considered similarity chromosome fitness values. optimal worst individuals were exchanged between two subpopulations...
The key to solving traffic congestion is the accurate speed forecasting. However, this difficult owing intricate spatial-temporal correlation of networks. Most existing studies either ignore correlations among distant sensors, or time-varying spatial features, resulting in inability extract and reliable features. To overcome these shortcomings, study proposes a new deep learning framework named gated graph convolutional network for long-term Firstly, generation method proposed, which uses...
Introduction: MET mutations have been linked with highly progressive clinical behaviors and poor prognosis in several tumor types. Savolitinib (also known as AZD6094, HMPL-504 volitinib) is a novel, potent, selective inhibitor investigated solid tumors.Methods: This an open-label, single arm, multi-center phase II study to evaluate the efficacy safety of savolitinib exon 14 skipping mutant PSC or other types NSCLC (NCT02897479). Eligible patients (pts) were diagnosed unresectable metastatic...
Objective Cerebral aneurysms are classified as severe cerebrovascular diseases due to hidden and critical onset, which seriously threaten life health. An effective strategy control intracranial is the regular diagnosis timely treatment by CT angiography (CTA) imaging technology. However, unpredictable patient movements make it challenging capture sub-millimeter-level ultra-high resolution images in a CTA scan. In order improve doctor's judgment, necessary clarity of cerebral aneurysm medical...
In this paper, the problem of integration and scheduling automated guided vehicles (AGVs) inspection machines in medical sample-testing laboratories is addressed with objective minimizing maximum turnaround time (TAT). A genetic algorithm (GA) a sliding window (STW) heuristic (GA-STW) proposed for studied while incorporating vehicle assignment (VAA) decoding operations. mathematical model AGV machine established, dual-coding method process used to solve model. VAA was assign suitable AGVs...
Single image reflection removal problem aims to divide a reflection-contaminated into transmission and image. It is canonical blind source separation highly ill-posed. In this paper, we present novel deep architecture called unfolded single network (DURRNet) which makes an attempt combine the best features from model-based learning-based paradigms therefore leads more interpretable architecture. Specifically, first propose optimization with transform-based exclusion prior then design...
Graph neural networks (GNNs) have been popularly used in analyzing graph-structured data, showing promising results various applications such as node classification, link prediction and network recommendation. In this paper, we present a new graph attention network, namely GIPA, for attributed data learning. GIPA consists of three key components: attention, feature propagation aggregation. Specifically, the component introduces multi-layer perceptron based multi-head to generate better...
e21157 Background: In ORIENT-11, first-line (1L) SPP improved progression-free survival (PFS), overall (OS), objective response rate (ORR) and duration of (DoR) compared with placebo plus pemetrexed-platinum (PPP) in pts AMnsqNSCLC. We report outcomes from ORIENT-11 (NCT03607539) for treated stage III (b & c) IV disease. Methods: Pts treatment-naïve locally AMnsqNSCLC without sensitizing EGFR/ALK alterations were randomly assigned to (n = 266) or PPP 131), stratified by PD-L1 expression,...
Medical image classification has been widely adopted in medical analysis. However, due to the difficulty of collecting and labeling data area, datasets are usually highly-imbalanced. To address this problem, previous works utilized class samples as prior for re-weighting or re-sampling but feature representation is still not discriminative enough. In paper, we adopt contrastive learning tackle long-tailed imbalance problem. Specifically, first propose category prototype adversarial...
Multi-user multiple-input multiple-output (MU-MIMO) beamforming design is typically formulated as a non-convex weighted sum rate (WSR) maximization problem that known to be NP-hard. This solved either by iterative algorithms, which suffer from slow convergence, or more recently using deep learning tools, require time-consuming pre-training process. In this paper, we propose low-complexity meta-learning based gradient descent algorithm. A meta network with lightweight architecture applied...