A5103193966- COVID-19 diagnosis using AI
- AI in cancer detection
- Brain Tumor Detection and Classification
- Analog and Mixed-Signal Circuit Design
- Advanced Battery Technologies Research
- Electrochemical Analysis and Applications
- ZnO doping and properties
- Image Retrieval and Classification Techniques
- Technology and Security Systems
- Power Systems Fault Detection
- Machine Fault Diagnosis Techniques
- Power Systems and Renewable Energy
- Advanced battery technologies research
- Catalytic Processes in Materials Science
- Physics of Superconductivity and Magnetism
- Islanding Detection in Power Systems
- Electric Vehicles and Infrastructure
- Transition Metal Oxide Nanomaterials
- Advanced Neural Network Applications
- Frequency Control in Power Systems
- Microgrid Control and Optimization
- Ionic liquids properties and applications
- Text and Document Classification Technologies
- CCD and CMOS Imaging Sensors
- Artificial Intelligence in Healthcare
Huazhong University of Science and Technology
2009-2024
Beijing Institute of Graphic Communication
2024
Brain tumor segmentation is a fundamental task and existing approaches usually rely on multi-modality magnetic resonance imaging (MRI) images for accurate segmentation. However, the common problem of missing/incomplete modalities in clinical practice would severely degrade their performance, fusion strategies incomplete brain are far from ideal. In this work, we propose novel framework named M 2 FTrans to explore fuse cross-modality features through modality-masked transformers under various...
Cross-silo federated learning (FL) enables decentralized organizations to collaboratively train models while preserving data privacy and has made significant progress in medical image classification. One common assumption is task homogeneity where each client access all classes during training. However, clinical practice, given a multi-label classification task, constrained by the level of knowledge prevalence diseases, institution may diagnose only partial categories, resulting...
The Green transportation is especially important as energy scarcity and air pollution from exhaust fumes fuel vehicles become more serious. Electric have the advantages of non-pollution, low noise, high efficiency simple structure, which will surely mainstream transportation. Brushless DC motor drive electric vehicles, it much efficient than traditional brushed motor, can reach 80-90% efficiency, in same volume output power, with higher power density. Due to its maintenance repair a...
Incomplete multi-modal image segmentation is a fundamental task in medical imaging to refine deployment efficiency when only partial modalities are available. However, the common practice that complete-modality data visible during model training far from realistic, as can have imbalanced missing rates clinical scenarios. In this paper, we, for first time, formulate such challenging setting and propose Preference-Aware Self-diStillatION (PASSION) incomplete under rates. Specifically, we...
Magnetizing inrush current in transformer has rich harmonics and large amplitude due to core saturation when the is energized. Large can have a negative impact on power equipment, system operation, electrical protection systems. The key of suppressing make accurate control circuit breaker's closing Angle which determined by opening angle. Therefore getting angle plays decisive effect current. Due interference signal, it hard determine actual However, process measuring residual magnetization...
This paper proposes a fault location method for distribution lines based on coupling parameter identification using two terminals data. In view of the problems insufficient deployment on-site phase selectors and inaccurate line parameters in medium voltage network, are identified by characteristic loop equation with pre-fault PMU recording On this basis, corresponding distance percentage is calculated genetic algorithm recorded data transient process three-phase synthetically. The weighting...
Replacing Zn metal with intercalated materials as the aqueous zinc-ion batteries (AZIBs) anode is an effective strategy to solve serious zinc dendrites problem. MXene-based derived combine advantages of both derivatives excellent electrochemical activity and MXene high electrical conductivity. Here, we report in situ for synthesizing H2Ti3O7−MXene from common Ti3C2 by one-step hydrothermal process simultaneous alkalization oxidation, followed ion-exchange reaction. It experimentally verified...
This work presents a novel capacitance-to-digital converter with adaptive parasitic compensation. It realizes wide measured range and high resolution compensation, oversampling noise shaping techniques. The circuit is fabricated in 0.18-μm CMOS process. In the post-layout simulation, it consumes 801μW under 3V supply 1kHz of sample frequency. design can detect 125pF by 0.1fF precision, which prove to be good candidate for dynamic CDC.