A5101512309- Advanced MIMO Systems Optimization
- Zeolite Catalysis and Synthesis
- Advanced Wireless Communication Technologies
- Millimeter-Wave Propagation and Modeling
- Advanced Decision-Making Techniques
- Medical Image Segmentation Techniques
- Advanced Sensor and Energy Harvesting Materials
- Impact of Light on Environment and Health
- Image Processing Techniques and Applications
- Building Energy and Comfort Optimization
- Evaluation Methods in Various Fields
- Catalytic Processes in Materials Science
- Image Retrieval and Classification Techniques
- Power System Optimization and Stability
- Advanced Neural Network Applications
- Advanced Materials and Mechanics
- Power Systems and Renewable Energy
- Optimal Power Flow Distribution
- Domain Adaptation and Few-Shot Learning
- Housing Market and Economics
- Cancer-related molecular mechanisms research
- Impact of AI and Big Data on Business and Society
- Superconducting and THz Device Technology
- Fluoride Effects and Removal
- Radio Astronomy Observations and Technology
University of Minnesota
2019-2024
Soochow University
2024
Xihua University
2024
Shanghai Jiao Tong University
2024
Twin Cities Orthopedics
2022-2024
China Agricultural University
2024
Tsinghua University
2024
Hunan University
2023
Shaanxi University of Science and Technology
2021-2023
Wuhan University of Science and Technology
2023
BackgroundMedical artificial intelligence (AI) has entered the clinical implementation phase, although real-world performance of deep-learning systems (DLSs) for screening fundus disease remains unsatisfactory. Our study aimed to train a clinically applicable DLS diseases using data derived from real world, and externally test model photographs collected prospectively settings in which would most likely be adopted.MethodsIn this national evidence study, we trained DLS, Comprehensive AI...
The emerging 5G services offer numerous new opportunities for networked applications. In this study, we seek to answer two key questions: i) is the throughput of mmWave predictable, and ii) can build "good" machine learning models prediction? To end, conduct a measurement study commercial in major U.S. city, focusing on as perceived by applications running user equipment (UE). Through extensive experiments statistical analysis, identify UE-side factors that affect performance quantify what...
Solving power flow (PF) equations is the basis of analysis, which important in determining best operation existing systems, performing security etc. However, PF can be out-of-date or even unavailable due to system dynamics and uncertainties, making traditional numerical approaches infeasible. To address these concerns, researchers have proposed data-driven solve problem by learning mapping rules from historical data. Nevertheless, prior suffer poor performance generalizability, overly...
Abstract Terahertz (THz) emission spectroscopy (TES) has emerged as a highly effective and versatile technique for investigating the photoelectric properties of diverse materials nonlinear physical processes in past few decades. Concurrently, research on two-dimensional (2D) experienced substantial growth due to their atomically thin structures, exceptional mechanical optoelectronic properties, potential applications flexible electronics, sensing, nanoelectronics. Specifically, these offer...
Well-functionalized electronic materials, such as silicon, in a stretchable format are desirable for high-performance wearable electronics. However, obtaining Si materials that meet the required stretchability of over 100% applications remains significant challenge. Herein, rational design strategy is proposed to achieve freestanding serpentine strips (FS-Si strips) with ultrahigh stretchability, fulfilling requirements. The self-supporting feature makes get rid excessive constraints from...
Transforming fluoride ions from reactants to catalysts in the silicon corrosion reaction enables their detection at high selectivity and sensitivity.
This work proposes a method of low-cost alkali etching to fabricate single-crystalline silicon frameworks, which will promote the research and application silicon-based imperceptible wearable electronics.
OP-Fe-FER-30 show better NH 3 -SCR activity than IM-Fe/SSZ-20 whether before or after hydrothermal aging treatments, which cause the collapse of zeolite framework, migration and aggregation iron species decease activity.
To expand the space applications of low-frequency sky-wave signals, Loran-C electromagnetic fields at low and high altitudes in earth–ionosphere waveguide are studied this letter. In order to consider influence earth's surface on one-hop different altitudes, wave-hop theory that used for field predictions ground is improved, equation based frequency domain further transformed into time domain. Then, proposed approach applied calculate 0–50 km under given models. The accuracy efficiency...
Performance of caching algorithms not only determines the quality experience for users, but also affects operating and capital expenditures cloud service providers. Today's production systems rely on heuristics such as LRU (least recently used) its variants, which work well certain types workloads, cannot effectively cope with diverse time-varying workload characteristics. While learning-based have been proposed to deal these challenges, they still impose assumptions about characteristics...
A centimeter-sized bearing fault probe based on dual-fiber Bragg grating vibration sensing is proposed. The can provide multi-carrier heterodyne measurements swept source optical coherence tomography technology and the synchrosqueezed wavelet transform method to obtain a wider frequency response range collect more accurate data. For sequential characteristics of signals, we propose convolutional neural network with long short-term memory transformer encoder. This proven in classification...
The [Fe, Al]-FER zeolites, which are prepared from the interzeolite conversion method (from MWW to FER topology) with high Al and Fe content (Si/Al of 8–18 Si/Fe 11–13), exhibit excellent catalytic performances hydrothermal tolerance for NH3-SCR reaction. process can be tuned by optimizing zeolite synthesis conditions such as gel composition, crystallization time, temperature, adding seeds. work provide a new strategy synthesize stability NOx removal.
In recent years, with the further breakthrough of artificial intelligence theory and technology, as well expansion Internet scale, recognition human emotions necessity for satisfying psychological needs in future technology development tendencies have been highlighted, addition to physical task accomplishment. Musical emotion classification is an important research topic intelligence. The key premise realizing music construct a musical model that conforms characteristics recognition....
A central prediction within the behavioral sciences asserts that agents are risk-seeking in domain of losses. This rests on experimental evidence did not involve realistic experience Here we tested whether this notion still holds if participants face To end, endowed with over $600 for at least one month and asked them to play gambles real, substantive, consequential losses spanned three orders magnitude. hierarchical Bayesian cognitive model comprising a latent mixture several competing...
Multiple Description Coding (MDC) is a promising error-resilient source coding method that particularly suitable for dynamic networks with multiple (yet noisy and unreliable) paths. However, conventional MDC video codecs suffer from cumbersome architectures, poor scalability, limited loss resilience, lower compression efficiency. As result, has never been widely adopted. Inspired by the potential of neural codecs, this paper rethinks design. We propose novel codec, NeuralMDC, demonstrating...
Grassland intelligence perception is conducive to adequate monitoring of grassland and saving human resources. However, road conditions are complex, vibration often disturbs image acquisition sensors. This study explores the influencing factors green vegetation imaging. According analysis variance, when collection equipment far away from vegetation, shooting an overhead angle, density low, obtained data has a higher proportion precise smaller blurred distorted images. proposed theoretical...
The conventional approaches for modeling dynamic systems are based on state-space methods using differential algebraic equations (DAEs). Such models not only require that the system dynamics can be precisely captured and expressed in mathematical equations, but also need detailed knowledge about parameters. Even when such DAEs available, no closed-form solutions numerical computationally expensive. As an example, modern power typically large complex networks comprising of hundreds or even...