A5100619427- Hydrological Forecasting Using AI
- Water Quality Monitoring Technologies
- Spectroscopy and Chemometric Analyses
- Hydrology and Watershed Management Studies
- Flood Risk Assessment and Management
- Pharmaceutical and Antibiotic Environmental Impacts
- 3D Shape Modeling and Analysis
- Nutrition and Health in Aging
- Identification and Quantification in Food
- Asphalt Pavement Performance Evaluation
- Recycling and utilization of industrial and municipal waste in materials production
- Rabies epidemiology and control
- Oral health in cancer treatment
- Traditional Chinese Medicine Analysis
- Smart Agriculture and AI
- Water-Energy-Food Nexus Studies
- Hydrology and Sediment Transport Processes
- Climate variability and models
- Microplastics and Plastic Pollution
- Oral and gingival health research
- Environmental DNA in Biodiversity Studies
- Functional Brain Connectivity Studies
- Bartonella species infections research
- Energy Load and Power Forecasting
- Medical Imaging and Analysis
Northeast Agricultural University
2024
Research Center for Eco-Environmental Sciences
2024
University of Chinese Academy of Sciences
2024
Beijing Institute of Technology
2018-2024
Wuhan University
2010-2023
Shandong University of Science and Technology
2023
Nanjing University of Information Science and Technology
2021
Tsinghua University
2021
Yale University
2021
University of Massachusetts Amherst
2019
In the task of using deep learning semantic segmentation model to extract water from high-resolution remote sensing images, multiscale feature and extraction have become critical factors that affect accuracy image classification tasks. A single-scale training mode will cause one-sided results, which can lead "reverse" errors imprecise detail expression. Therefore, fusing features for pixel-level is key achieving accurate segmentation. Based on this concept, paper proposes a scheme achieve...
To assess the status of, and factors associated with, residents' knowledge, attitudes, practices (KAP) related to air pollution respiratory health of children in Shanghai, we conducted a cross-sectional survey. Demographic with knowledge were identified by multiple logistic regressions. The questionnaires completed 972 participants, half from Shanghai Children Hospital other Jiading communities. Half participants' scores attitudes equal or greater than 8.0 on 9-point scale, over 75%...
Bartonella species are facultative intracellular bacteria and recognized worldwide as emerging zoonotic pathogens. were isolated or identified by polymerase chain reaction (PCR) in bats their ectoparasites worldwide, whereas the association between them was scarce, especially Asia. In this study, a retrospective analysis with frozen samples carried out to identify genetic diversity of investigate relationships ectoparasites. Bats (bat flies bat mites) collected from caves Hubei Province,...
Positron emission tomography (PET) imaging with various tracers is increasingly used in Alzheimer's disease (AD) studies. However, access to PET scans using new or less-available sophisticated synthesis and short half-life isotopes may be very limited. Therefore, it of great significance interest AD research assess the feasibility generating synthetic images from image another common tracer, particular
The corrupted information in training samples is an important factor affecting the accuracy and generalizability of machine learning models. Due to extremely high memory capacity deep models, interference excessive makes model prone bad generalization behavior. This paper proposes a method estimating sample quality using value calculated by loss function process gradient descent optimization. includes variation degree algorithm analysis algorithm. provides basis for determining intervention...
Chinese herbal medicine is an important part of traditional (TCM). With developing medicine, the usage growing rapidly. It essential to identify correctly since used treat disease. However, identifying a hard task because lots with different properties displays similar appearance, such as Radix StephaniaeTetrandrae and Paeoniae Alba. Traditional methods classifying are low-efficiency rely on professional medical knowledge. Machine learning can reduce need for knowledge in some fields due its...
Realizing the green recycling of sludge is an important link to effectively solve problem disposal. In this paper, sewage (SS) and rice husk (RH) were utilized as raw materials in preparing novel ceramsite (SRC) for treatment lead-containing wastewater, its adsorption mechanism was explored. The results showed that optimal preparation conditions 40% RH + 60% SS mixture, a sintering temperature 1190 °C, time 20 min. basic properties SRC met Chinese artificial filter material standards water...
The Haihe River Basin (HRB), located in northern China with a drainage area of 318,200 km, is one the most developed regions China. With rapid population growth and economic development, combined problems water shortage groundwater overpumping significantly constrain sustainable development this area. In order to strengthen unified management surface water, we hydrologic modeling interaction by coupling SWAT (for simulation) MODFLOW simulation). newly framework reasonably captured...
Current deep learning models for dynamics forecasting struggle with generalization. They can only forecast in a specific domain and fail when applied to systems different parameters, external forces, or boundary conditions. We propose model-based meta-learning method called DyAd which generalize across heterogeneous domains by partitioning them into tasks. has two parts: an encoder infers the time-invariant hidden features of task weak supervision, forecaster learns shared entire domain. The...
As tomato leaves are often attacked by various microorganisms, pests and bacterial diseases, the yield of is seriously reduced. Accurate timely identification leaf diseases great significance to reduce farmers' economic losses. Ensemble learning as a combinatorial optimization method, which can improve generalization ability model stability, widely used in field plant disease identification. However, commonly ensemble methods such majority voting, weighted averaging, etc. do not consider...
The crack feature is the key to recognition in concrete surface detection systems based on machine vision. As “essence” of crack, skeleton shows good stability. Therefore, a extraction method inflection point proposed this paper. main processing steps include image preprocessing, extraction, and recognition. In extracting skeleton, initially extracted using morphological operations. After identifying endpoints neighborhood distribution points, burrs are tracked until branching as starting...
Necessity, history, actuality, problems and countermeasures, industrial application prospect of seawater used as cycling cooling water are analysed.When is water, waste discharge quantity intake small, engineering investment operation cost low, it more reasonable in economy.It favour saving fresh resources, easing tensions protecting marine environment coastal areas, keeping ecological balance, but the corrosion, scaling biological control system running, including salt deposition, fog...
As an important research direction of the application computer science in medical field, automatic generation technology radiology report has attracted wide attention academic community.Because proportion normal regions images is much larger than that abnormal regions, words describing diseases are often masked by other words, resulting significant feature loss during calculation process, which affects quality generated reports.In addition, huge difference between visual features and...