A5100318314- EEG and Brain-Computer Interfaces
- Video Surveillance and Tracking Methods
- Emotion and Mood Recognition
- ECG Monitoring and Analysis
- Human Pose and Action Recognition
- Gait Recognition and Analysis
- Face recognition and analysis
- Neural dynamics and brain function
- Functional Brain Connectivity Studies
- Advanced Neural Network Applications
- Gaze Tracking and Assistive Technology
- Non-Invasive Vital Sign Monitoring
- Blind Source Separation Techniques
- Cardiac Fibrosis and Remodeling
- Cardiomyopathy and Myosin Studies
- Image Enhancement Techniques
- Viral Infections and Immunology Research
- Reinforcement Learning in Robotics
- Network Security and Intrusion Detection
- Atrial Fibrillation Management and Outcomes
- Software-Defined Networks and 5G
- Phonocardiography and Auscultation Techniques
- Heart Rate Variability and Autonomic Control
- Neural Networks and Applications
- Cardiac Arrhythmias and Treatments
Southeast University
2016-2025
Affiliated Hospital of Guizhou Medical University
2022-2025
Guiyang Medical University
2022-2025
Beijing Tongren Hospital
2021-2024
Capital Medical University
2021-2024
Huazhong University of Science and Technology
2016-2024
Beijing University of Posts and Telecommunications
2024
Second Affiliated Hospital of Nanjing Medical University
2021-2024
Kunming University of Science and Technology
2023
First People's Hospital of Yunnan Province
2023
Existing person re-identification (re-id) methods either assume the availability of well-aligned bounding box images as model input or rely on constrained attention selection mechanisms to calibrate misaligned images. They are therefore sub-optimal for re-id matching in arbitrarily aligned potentially with large human pose variations and unconstrained auto-detection errors. In this work, we show advantages jointly learning feature representation a Convolutional Neural Network (CNN) by...
Most existing person re-identification (re-id) methods require supervised model learning from a separate large set of pairwise labelled training data for every single camera pair. This significantly limits their scalability and usability in real-world scale deployments with the need performing re-id across many views. To address this problem, we develop novel deep method transferring information an dataset to new unseen (unlabelled) target domain without any domain. Specifically, introduce...
Existing person re-identification (re-id) methods rely mostly on either localised or global feature representation. This ignores their joint benefit and mutual complementary effects. In this work, we show the advantages of jointly learning local features in a Convolutional Neural Network (CNN) by aiming to discover correlated different context. Specifically, formulate method for selection losses designed optimise re-id when using generic matching metrics such as L2 distance. We design novel...
Emotion classification based on brain-computer interface (BCI) systems is an appealing research topic. Recently, deep learning has been employed for the emotion classifications of BCI and compared to traditional methods improved results have obtained. In this paper, a novel neural network proposed using EEG systems, which combines Convolutional Neural Network (CNN), Sparse Autoencoder (SAE) Deep (DNN) together. network, features extracted by CNN are sent SAE encoding decoding at first. Then...
Atrial fibrillation (AF) is the most common persistent cardiac arrhythmia. This study aimed to estimate its prevalence and explore associated factors in adults aged 18 years or older China.Study data were derived from a national sample July 2020 September 2021. Participants recruited using multistage stratified sampling method twenty-two provinces, autonomous regions, municipalities China. AF was determined based on history of diagnosed electrocardiogram results.A total 114,039 respondents...
Existing person re-identification (re-id) methods either assume the availability of well-aligned bounding box images as model input or rely on constrained attention selection mechanisms to calibrate misaligned images. They are therefore sub-optimal for re-id matching in arbitrarily aligned potentially with large human pose variations and unconstrained auto-detection errors. In this work, we show advantages jointly learning feature representation a Convolutional Neural Network (CNN) by...
Acute myocardial infarction (AMI) has become a major cause of hospitalization and mortality in China. There been limited data to date available characterize AMI presentation, contemporary patterns medical care, outcomes China.The CAMI Registry is national project with the objectives timely obtain real-world knowledge about patients provide platform for clinical research, guide preventive measures care quality improvement efforts registry prospective, nationwide, multicenter observational...
The issue of electroencephalogram (EEG)-based emotion recognition has great academic and practical significance. Currently, there are numerous research trying to address this in the literature. Particularly, transfer learning gradually become a new methodological trend for company with popularity deep learning. Motivated by capturing panorama, summarizing technological essence, forecasting advancement tendency EEG-based recognition, article contributes review work. This work mainly includes...
Insect pest management is one of the main ways to improve crop yield and quality in agriculture it can accurately timely detect insect pests, which great significance agricultural production. In past, most detection tasks relied on experience agricutural experts, time-consuming, laborious subjective. rencent years, various intelligent methods have emerged for detection. This paper employs three frontier Deep Convolutional Neural Network (DCNN) models—Faster-RCNN, Mask-RCNN Yolov5, efficient...
Myocardial fibrosis is the characteristic pathology of diabetes-induced cardiomyopathy. Therefore, an in-depth study cardiac heterogeneity and cell-to-cell interactions can help elucidate pathogenesis diabetic myocardial identify treatment targets for this disease. In study, we investigated intercellular communication drivers in mouse heart with high-fat-diet/streptozotocin-induced diabetes at single-cell resolution. Intercellular protein–protein interaction networks fibroblasts macrophages,...
Abstract Background Autoimmune diseases (ADs) present significant health challenges globally, especially among adolescents and young adults (AYAs) due to their unique developmental stages. Comprehensive analyses of burden are limited. This study leverages the Global Burden Disease (GBD) 2021 data assess global, regional, national trends major ADs AYAs from 1990 2021. Methods Utilizing Study for individuals aged 15–39 years, we employed a direct method age standardization calculate estimates...
Electroencephalogram (EEG)-based emotion recognition is of great significance for aiding in clinical diagnosis, treatment, nursing and rehabilitation. Current research on this issue mainly focuses utilizing various network architectures with different types neurons to exploit the temporal, spectral, or spatial information from EEG classification. However, most studies fail take full advantage useful Temporal-Spectral-Spatial (TSS) signals. In paper, we propose a novel effective Fractal Spike...
Agriculture is important for ecology. The early detection and treatment of agricultural crop diseases are meaningful challenging tasks in agriculture. Currently, the identification plant relies on manual detection, which has disadvantages long operation time low efficiency, ultimately impacting yield quality. To overcome these disadvantages, we propose a new object method named “Plant Leaf Detection transformer with Improved deNoising anchOr boxes (PL-DINO)”. This incorporates Convolutional...
Macrophages play an important role in the development of cardiac fibrosis. However, roles different macrophage subtypes fibroblast (CF) activation and fibrosis are unknown.
In recent years, there has been a growing research interest in using deep learning to resolve the issue of electroencephalogram (EEG)-based emotion recognition. Current emphasizes exploiting useful information from each single EEG channel or individual set multichannel EEG, but overlooks correlation among different sets. To explore such discriminative information, we propose novel and effective method, "trainable adjacency relation driven graph convolutional network (TARDGCN)," which...
Physiological and pathological information within electrocardiogram (ECG) is crucial for the diagnosis of heart diseases. Computer-aided ECG signals has drawn growing research attention up to date. Automatic analysis mainly includes signal denoising, wave detection, heartbeat classification. These three issues are relevant that denoising can help attenuate noises accentuate typical waves in detection locate acquire diagnostically valuable heartbeats based on these The wavelet-based methods...
Electroencephalogram (EEG)-based emotion recognition has gradually become a research hotspot. However, the large distribution differences of EEG signals across subjects make current stuck in dilemma. To resolve this problem, article, we propose novel and effective method, Multi-Source Feature Representation Alignment Network (MS-FRAN). The effectiveness proposed method mainly comes from three new modules: Wide Extractor (WFE) for feature learning, Random Matching Operation (RMO) model...
Abstract Background Evaluation of the primary etiologic agents that cause aseptic meningitis outbreaks may provide valuable information regarding prevention and management meningitis. An outbreak occurred from May to June, 2012, in Guangdong Province, China. In order determine agent, CSF specimens 121 children hospitalized for at Luoding People’s Hospital Province were tested virus isolation identification. Results Enterovirus RNA was positive 62.0% sspecimens by real-time polymerase chain...
To reduce the high mortality rate among heart patients, electrocardiogram (ECG) beat classification plays an important role in computer aided diagnosis system, but this issue is challenging because of complex variations data. Since ECG data lie on high-dimension manifold, we propose a novel method, named “local deep field”, purpose capturing devil details such manifold. This method learns different models within local manifold charts. Local regionalization can help focus particularity...