A5009416262- Parasites and Host Interactions
- Liver Disease and Transplantation
- Liver Disease Diagnosis and Treatment
- Railway Engineering and Dynamics
- Vehicle License Plate Recognition
- Infrastructure Maintenance and Monitoring
- Eicosanoids and Hypertension Pharmacology
- Dental Radiography and Imaging
- Pancreatitis Pathology and Treatment
- Bacillus and Francisella bacterial research
- Robotics and Sensor-Based Localization
- Forensic Anthropology and Bioarchaeology Studies
- Macrophage Migration Inhibitory Factor
- Anomaly Detection Techniques and Applications
- Medical Image Segmentation Techniques
- Advanced Data Processing Techniques
- Ginkgo biloba and Cashew Applications
- Remote Sensing and LiDAR Applications
- Cancer-related molecular mechanisms research
- Natural Antidiabetic Agents Studies
- Domain Adaptation and Few-Shot Learning
- Phytochemicals and Antioxidant Activities
- Multimodal Machine Learning Applications
- Anatomy and Medical Technology
- 3D Surveying and Cultural Heritage
Jiangsu Academy of Agricultural Sciences
2024-2025
Central South University
2023-2024
Third Xiangya Hospital
2023-2024
Hunan Provincial Center for Disease Control and Prevention
2024
Hunan Institute of Microbiology
2023
Institute for Infocomm Research
2018-2021
Agency for Science, Technology and Research
2018-2021
National University of Singapore
2015-2018
Abstract Background The interactions between blueberry leaf polyphenols (BLPs) and digestive enzymes were analyzed using multiple techniques to gain insights into their inhibitory effects on enzyme kinetics modes of action. Results 3‐ O ‐Caffeoylquinic acid (3‐CQA) was the most abundant compound identified. Quercetin (QR) exhibited strongest activity against α ‐amylase ( ‐AMY) ‐glucosidase ‐GLU). BLP extracts acted as typical mixed‐type inhibitors for both enzymes, showing stronger...
Regular inspection of rail valves and engines is an important task to ensure safety efficiency railway networks around the globe. Over past decade, computer vision pattern recognition based techniques have gained traction for such defect detection tasks. An automated end-to-end trained system can potentially provide a low-cost, high throughput, cheap alternative manual visual these components. However, systems require huge amount defective images understand complex defects. In this paper,...
ABSTRACT Hepatic ischaemia–reperfusion (I/R) injury is a frequent and nearly inevitable pathophysiological process without widely accepted effective therapy. Soluble egg antigen (SEA) of Schistosoma japonicum ( S. ) the main mediators capable regulating immunological activities has received increased attention in immune‐mediated diseases. But its role hepatic I/R not been well defined. This study aimed to elucidate whether SEA protects liver against explore underlying mechanism. After...
There are two challenging problems in applying standard Deep Neural Networks (DNNs) for incremental learning with a few examples: (i) DNNs do not perform well when little training data is available; (ii) suffer from catastrophic forgetting used class learning. To simultaneously address both problems, we propose Meta Module Generation (MetaMG), meta-learning method that enables module generator to rapidly generate category examples scalable classification network recognize new category. The...
Reconstruction of skulls from defective models is a very important and challenging task in craniofacial surgery, forensics, anthropology. Existing methods typically reconstruct approximating surfaces that regard corresponding points on the target skull as soft constraints, thus incurring non-zero error even for non-defective parts high overall reconstruction error. This paper proposes novel method non-rigidly registers an interpolating surface regards hard achieving low To overcome...
We propose a novel method and system to prevent indiscriminate parking of dockless shared bicycles using location-based geo-fencing image-based place identification. The is used define the approximate regions for different types bicycle regulations. identification uses based on deep Convolutional Neural Network (DCNN) automatically identify designated places from photos captured by cyclist mobile phone. Combining these two modalities, can be restricted in zones various environments....
Background: Schistosomiasis has persisted in China for over a thousand years. Despite decades of efforts, significant progress been made prevention and control. However, certain regions, like Qingshan Island, still pose challenges, reflecting 'dark corner' the final stages control, hindering goal elimination. Methods: Cross-sectional research, conducted on Island China, investigated burden japonicum among villagers, involved 133 residents by questionnaire methods. Data were analyzed...
Abstract This study intends to use the basic information and blood routine of schistosomiasis patients establish a machine learning model for predicting liver fibrosis. We collected medical records Schistosoma japonicum admitted hospital in China from June 2019 2022. The method was screen out key variables six different algorithms were used prediction models. Finally, optimal compared based on AUC, specificity, sensitivity other indicators further modeling. interpretation shown by using SHAP...
Schistosoma japonicum (S. japonicum) is the main species of prevalent in China. Myeloid-derived suppressor cells (MDSCs) are important immunoregulatory and generally expand parasite infection, but there little research relating to MDSCs infection.
Skull reconstruction is an important and challenging task in craniofacial surgery planning, forensic investigation anthropological studies. Existing methods typically reconstruct approximating surfaces that regard corresponding points on the target skull as soft constraints, thus incurring non-zero error even for non-defective parts high overall error. This paper proposes a novel geometric method non-rigidly registers interpolating reference surface regards hard achieving low To overcome...
An important problem in artificial intelligence is to develop an efficient system that can adapt new knowledge incremental manner without forgetting previously learned knowledge. Although Convolutional Neural Networks (CNNs) are good at learning strong classifier and discriminative features, CNNs not perform well due the catastrophic retraining process. In this paper, we propose a novel yet extremely simple approach enhance property of features for learning. We build network universal...
One pivot challenge for image anomaly (AD) detection is to learn discriminative information only from normal class training images. Most reconstruction based AD methods rely on the capability of error. This heuristic as unsupervised without incorporating normal-class-specific information. In this paper, we propose an method called dual deep networks decomposition (DDR-ID). The are trained by jointly optimizing three losses: one-class loss, latent space constrain loss and loss. After...
Abstract Schistosomiasis is a chronic parasitic disease, which affects the quality of daily life patients and imposes huge burden on society. Hepatic fibrosis in response to continuous insult eggs liver significant cause morbidity mortality. However, mechanisms hepatic schistosomiasis are largely undefined. The purpose our study detect indicator schistosomiasis. A total 488 with japonica were enrolled study. divided into two groups according ultrasound examination, could indicate unique...
Schistosomiasis is a chronic parasitic disease, which affects the quality of daily life patients and imposes huge burden on society. Hepatic fibrosis in response to continuous insult eggs liver significant cause morbidity mortality. However, mechanisms hepatic schistosomiasis are largely undefined. The purpose our study detect indicator schistosomiasis. A total 488 with japonica were enrolled study. divided into two groups according ultrasound examination, could indicate unique reticular...
Abstract This study intends to use the basic information and blood routine of schistosomiasis patients establish a machine learning model for predicting liver fibrosis. We collected medical records Schistosoma japonicum admitted hospital in China from June 2019 2022. The method was screen out key variables six different algorithms were used prediction models. Finally, optimal compared based on AUC, specificity, sensitivity other indicators further modeling. interpretation shown by using SHAP...
Regular inspection of rail valves and engines is an important task to ensure the safety efficiency railway networks around globe. Over past decade, computer vision pattern recognition based techniques have gained traction for such defect detection tasks. An automated end-to-end trained system can potentially provide a low-cost, high throughput, cheap alternative manual visual these components. However, systems require huge amount defective images understand complex defects. In this paper,...