A5115604796- Advanced Neural Network Applications
- Reinforcement Learning in Robotics
- Industrial Vision Systems and Defect Detection
- Robotics and Sensor-Based Localization
- Nuclear reactor physics and engineering
- Nuclear Materials and Properties
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and LiDAR Applications
- Evolutionary Algorithms and Applications
- Advanced X-ray and CT Imaging
- Assembly Line Balancing Optimization
- Silicon Carbide Semiconductor Technologies
- Advanced Optical Sensing Technologies
- Medicinal Plants and Neuroprotection
- COVID-19 diagnosis using AI
- Advanced Multi-Objective Optimization Algorithms
- Gut microbiota and health
- Gestational Diabetes Research and Management
- Geophysics and Gravity Measurements
- Brain Tumor Detection and Classification
- Inertial Sensor and Navigation
- Brake Systems and Friction Analysis
- Control and Dynamics of Mobile Robots
- Advanced Manufacturing and Logistics Optimization
- Multi-Agent Systems and Negotiation
Hangzhou Medical College
2025
Shanghai Jiao Tong University
2022-2025
China Medical University
2025
Dalian University of Technology
2025
Liaoning Cancer Hospital & Institute
2025
Zhejiang University
2025
Chinese Academy of Fishery Sciences
2024
Ministry of Agriculture and Rural Affairs
2024
Affiliated Eye Hospital of Wenzhou Medical College
2024
Wenzhou Medical University
2024
In the conventional digital peak current (DPC) predictive controlled switching dc–dc converters, duty ratio in each cycle is usually calculated and predicted based on status of previous cycles. Therefore at least one-switching-cycle time delay exists such a control technique, which significantly degrades performance digitally converters. To eliminate this to improve performance, an improved (IDPC) technique for converters then proposed studied study. The laws DPC IDPC with different pulse...
Obesity is currently one of the most important challenges to public health worldwide. Acupuncture has been widely used treat obesity. However, whether acupuncture regulates intestinal innate immunity via microbiota against obesity remains be elucidated. In this study, electroacupuncture (EA) effectively reduced body weight and fat accumulation in obese mice persistently fed a high-fat diet. Full-length 16S rDNA sequencing showed dysbiotic cecum mice. The composition function cecal were...
Objective The process of glycolysis from blood collection to centrifugation impacts the diagnosis gestational diabetes mellitus (GDM). However, specific characteristics working environment in China and its influence on GDM still need be clarified.
<title>Abstract</title> <bold>Introduction</bold> Patients with POLE/POLD1 gene mutation and dMMR/MSI-H are the dominant groups immune benefit may from immunotherapy. At present all guidelines recommend observation or chemotherapy based on stage risk factors for postoperative adjuvant treatment of gastric adenocarcinoma colorectal cancer. There is no definitive conclusion as to whether checkpoint inhibitors standard more appropriate.This study aims explore optimal mode locally advanced...
Background The visceral adiposity index (VAI) reliably measures body fat distribution and related dysfunctions. However, its association with sleep disorders among US adults remains unclear. Methods This study analyzed cross-sectional data from the 2005 to 2018 National Health Nutrition Examination Survey (NHANES) for aged 18 older. We used multivariable logistic regression evaluate between VAI applied restricted cubic splines assess potential non-linear relationships. Additionally, subgroup...
The semantic segmentation of agricultural aerial images is very important for the recognition and analysis farmland anomaly patterns, such as drydown, endrow, nutrient deficiency, etc. Methods general Fully Convolutional Networks can extract rich features, but are difficult to exploit long-range information. Recently, vision Transformer architectures have made outstanding performances in image tasks, transformer-based models not been fully explored field agriculture.Therefore, we propose a...
ObjectiveTo develop deep learning methods with high accuracy for segmenting irregular corneas and detecting the tear fluid reservoir (TFR) boundary under scleral lens. Additionally, this study aims to provide a publicly available cornea lens OCT dataset, including manually labeled layer masks training validation of segmentation algorithms. This introduces ScLNet, dataset comprising Scleral Lens (ScL) optical coherence tomography (OCT) images annotations, multi-task network designed achieve...
Abstract. Determining the attitude of satellite at time imaging then establishing mathematical relationship between image points and ground is essential in high-resolution remote sensing mapping. Star tracker insensitive to high frequency variation due measure noise jitter, but low motion can be determined with accuracy. Gyro, as a short-term reference satellite’s attitude, sensitive change, existence gyro drift integral error, determination error increases time. Based on opposite...
Abstract Objective . Decomposition methods are efficient to decode steady-state visual evoked potentials (SSVEPs). In recent years, the brain–computer interface community has also been developing deep learning networks for decoding SSVEPs. However, there is no clear evidence that current models outperform decomposition on SSVEP tasks. Many studies lacked comparison with state-of-the-art in a fair environment. Approach This study proposed novel network design motivated by works of methods....
What is the essence of Chinese Calligraphy beauty? It most important and difficult question in history art. The explorations that ancient artists had made on this can be classified into three main schools: first school emphasizes objective manifestation form, which means calligraphy beauty natural its form; second subjective intention, emotion; third combination subjectivity objectivity. In my opinion, it necessary to combine objectivity when we understand beauty. To it, however, cannot just...
Abstract Offline reinforcement learning (RL) aims to create policies for sequential decision-making using exclusively offline datasets. This presents a significant challenge, especially when attempting accomplish multiple distinct goals or outcomes within given scenario while receiving sparse rewards. Prior methods advantage weighting goal-conditioned improve monotonically. However, they still face challenges from distribution shift and multi-modality that arise due conflicting ways reach...
Traditional offline reinforcement learning methods predominantly operate in a batch-constrained setting. This confines the algorithms to specific state-action distribution present dataset, reducing effects of distributional shift but restricting algorithm greatly. In this paper, we alleviate limitation by introducing novel framework named \emph{state-constrained} learning. By exclusively focusing on dataset's state distribution, our significantly enhances potential and reduces previous...