- Photoacoustic and Ultrasonic Imaging
- Soft Robotics and Applications
- Image Processing Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Simulation and Modeling Applications
- Thermography and Photoacoustic Techniques
- Remote-Sensing Image Classification
- Advanced Graph Neural Networks
- Lymphatic System and Diseases
- Human Mobility and Location-Based Analysis
- Medical Imaging and Analysis
- Radiative Heat Transfer Studies
- Big Data Technologies and Applications
- Fire Detection and Safety Systems
- Cardiovascular and Diving-Related Complications
- DNA and Biological Computing
- Topic Modeling
- Optical Systems and Laser Technology
- Infrastructure Maintenance and Monitoring
- Risk and Safety Analysis
- Distributed Control Multi-Agent Systems
- Image and Object Detection Techniques
- Sparse and Compressive Sensing Techniques
- Heat Transfer Mechanisms
- Fire dynamics and safety research
Civil Aviation University of China
2025
Technical University of Munich
2024
Sichuan Normal University
2024
Beijing Institute of Power Machinery (China)
2024
Tianjin University
2022-2024
Tsinghua University
2024
University of South China
2023
Shenzhen Institutes of Advanced Technology
2023
Chinese Academy of Sciences
2023
Anhui University of Technology
2023
Negative pressure (NP) therapy is effective in managing chronic lymphedema of the extremities. However, seal formation general head (GH) can fail due to interspace between lip and irregular skin surface on limbs before suction, resulting inefficiency therapy, prolonging time required for physiotherapy, increasing workload physiotherapists. In this letter, we present a bio-inspired (BIH) that uses adaptive control NP unstructured surfaces. Its designed with soft material inspired by bloodworm...
Beyond pure text, a substantial amount of knowledge is stored in tables. In real-world scenarios, user questions often require retrieving answers that are distributed across multiple GraphRAG has recently attracted much attention for enhancing LLMs' reasoning capabilities by organizing external to address ad-hoc and complex questions, exemplifying promising direction cross-table question answering. this paper, the current gap available data, we first introduce multi-table benchmark,...
Robot-assisted endovascular intervention has the potential to reduce radiation exposure surgeons and enhance outcomes of interventions. However, success safety interventions depend on surgeons' ability accurately manipulate tools such as guidewire catheter perceive their when cannulating patient's vessels. Currently, existing interventional robots lack a haptic system for accurate force feedback that can rely on. In this paper, haptic-enabled robot was developed. We proposed dynamic...
Optical performance evaluation is a critical process in the production of collimating lenses. However, current visual inspection lens light-spot images inefficient and prone to fatigue. Intelligent detection based on machine vision deep learning can improve efficiency accuracy. In this study, dual-branch structure model proposed for optical evaluation, image dataset built, containing 9000 with corresponding labels. Experimental results show that achieves accurate classification evaluation....
In order to better focus on ability cultivation and stimulate students' innovative abilities, this paper proposes a teaching model based the Educoder platform for machine learning courses. This analyzes drawbacks of traditional teaching, combines needs as discipline, through mode reform, combined with platform, demonstrates operational application it in teaching. It not only greatly improves interest, but also allows students truly become main body learning, consequently yielding substantial...
With the rapid development of mobile Internet technology, number customers has increased dramatically, and applications have become widespread. Using big data generated by network activities to analyze customer behavior achieve accurate marketing a new demand in field Internet. In this paper, algorithm is used characteristics match classification, obtain requirements. The extraction method based on obtained. Through analysis characteristics, intelligent travel route recommendation interests...
Flexible needles have been widely used in minimally invasive surgeries, especially percutaneous interventions. Among the interventions, tip position of curved needle is very important, since it directly affects success surgeries. In this paper, we present a method to estimate long-curved by using 2D transverse ultrasound images from robotic system. Ultrasound first detect cross section long-flexible needle. A new imaging approach proposed based on selection numbers pixels with higher gray...
Photoacoustic tomography technology is a new non-invasive, non-ionizing biomedical imaging method. This combines the high contrast of optical and high-resolution characteristics ultrasound imaging, which can obtain images in deeper tissues. In recent years, it has developed rapidly won widespread attention. Traditional sampling method must follow Nyquist theorem, wastes lot sensing time storage space. order to improve efficiency, compressed (CS) theory used collect process photoacoustic...
Hyperspectral image (HSI) clustering is gaining considerable attention owing to recent methods that overcome the inefficiency and misleading results from absence of supervised information. Contrastive learning excel at existing pixel level super HSI tasks. The pixel-level contrastive method can effectively improve ability model capture fine features but requires a large time overhead. utilizes homogeneity reduces computing resources; however, it yields rough classification results. To...
Hypergraphs naturally arise when studying group relations and have been widely used in the field of machine learning. There has not a unified formulation hypergraphs, yet recently proposed edge-dependent vertex weights (EDVW) modeling is one most generalized methods i.e., existing hypergraphs can be formulated as EDVW without any information loss to best our knowledge. However, relevant algorithmic developments on remain nascent: compared spectral graph theories, formulations are incomplete,...
As a functional imaging technique, photoacoustic has the ability to produce high contrast and resolution image. However, microscopy shows disadvantage of limited depth field, due strong focusing process. Only some part whole image is clear while other information might be lost or misleading. To achieve large volumetric photoacoustic, we propose depth-of-field expansion motivated multi-focus fusion algorithm which fuses from different focal positions into single one obtain all-in-focus this...
Photoacoustic microscopy suffers from limited depth-of-field due to the strong focus of laser beam, which implies that only some part imaging result is clear. Such shortage limitsthe further application this powerful technique, problem we hope address. In work, consider solve target through information fusion method. By fusing multi-focus 3D photoacoustic images, a large volumetric and high resolution can be obtained. However, task signal different general 2-D image problem. The core...