A5100778754- Privacy-Preserving Technologies in Data
- Cryptography and Data Security
- Cooperative Communication and Network Coding
- Adversarial Robustness in Machine Learning
- Stochastic Gradient Optimization Techniques
- Geochemistry and Geologic Mapping
- Image Enhancement Techniques
- Chaos-based Image/Signal Encryption
- Machine Learning in Healthcare
- Advanced Wireless Communication Techniques
- Advanced Decision-Making Techniques
- Artificial Intelligence in Healthcare and Education
- Geological Modeling and Analysis
- Face recognition and analysis
- Video Surveillance and Tracking Methods
- Advanced Malware Detection Techniques
- Advanced Wireless Communication Technologies
- Soil Geostatistics and Mapping
- Mobile Crowdsensing and Crowdsourcing
- Full-Duplex Wireless Communications
- Advanced Optical Sensing Technologies
- Biometric Identification and Security
- Software-Defined Networks and 5G
- Color Science and Applications
- Data Quality and Management
Huazhong University of Science and Technology
2025
RMIT University
2019-2025
First Affiliated Hospital Zhejiang University
2025
Zhejiang University
2025
MIT University
2023-2024
Northwest University
2017-2024
China University of Geosciences (Beijing)
2019-2024
Shanghai Jiao Tong University
2016-2024
XinHua Hospital
2024
Weihai Science and Technology Bureau
2024
Neural network (NN) inference services enrich many applications, like image classification, object recognition, facial verification, and more. These NN are increasingly becoming an essential offering from cloud computing providers, where end-users' data offloaded to the for under a customized model. However, current cloud-based operate on clear inputs models, raising paramount privacy concerns. Individual user may contain private information that should always remain confidential. Meanwhile,...
The advances in machine learning have revealed its great potential for emerging mobile applications such as face recognition and voice assistant. Models trained via a Neural Network (NN) can offer accurate efficient inference services users. Unfortunately, the current deployment of service encounters privacy concerns. Directly offloading model to device violates owner, while feeding user input compromises privacy. To address this issue, we propose Leia, lightweight cryptographic NN system at...
The substantial progress in privacy-preserving machine learning (PPML) facilitates outsourced medical computer-aided diagnosis (MedCADx) services. However, existing PPML frameworks primarily concentrate on enhancing the efficiency of prediction services, without exploration into diverse services such as segmentation. In this paper, we propose <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MedShield</monospace> , a pioneering cryptographic...
Background: Integrating comprehensive information on hepatocellular carcinoma (HCC) is essential to improve its early detection. We aimed develop a model with multi-modal features (MMF) using artificial intelligence (AI) approaches enhance the performance of HCC Materials and methods: A total 1,092 participants were enrolled from 16 centers. These allocated into training, internal validation, external validation cohorts. Peripheral blood specimens collected prospectively subjected mass...
Three-dimensional Mineral Prospectivity Mapping (3DMPM) is an innovative approach to mineral exploration that combines multiple geological data sources create a three-dimensional (3D) model of deposit. It provides accurate representation the subsurface can be used identify areas with potential. These 3D models are typical source for prospective modeling. Geological sets from construct models. Since in practice there significant imbalance ratio mineralized non-mineralized classes,...
Medical time series data analytics based on dynamic warping (DTW) greatly benefits modern medical research. Driven by the distributed nature of data, collaboration multiple healthcare institutions is usually necessary for a sound conclusion. Among others, typical use case disease screening public health, where wish to collaboratively detect over their joint datasets patients whose records have similar features given query samples. However, sharing faces critical privacy obstacles with...
Similarity-oriented services serve as a foundation in wide range of data analytic applications such machine learning, target advertising, and real-time decisions. Both industry academia strive for efficient scalable similarity discovery querying techniques to handle massive, complex records the real world. In addition performance, security privacy become an indispensable criterion quality service due progressively increased breaches. To address this serious concern, paper, we propose...
An exception-related bug is a kind of program which causes exceptions. During software maintenance, when programmers repair bugs, they typically analyze thrown exceptions to understand the root such bugs. When encountering unfamiliar exceptions, often refer online forum threads (e.g. StackOverflow) how fix them. Although some general search engines are available and research tools proposed, insufficient recommend for bugs from large-scale resources. In this paper, we propose an approach,...
The Jiaodong Peninsula hosts the main large gold deposits and was first production area in China; multisource multiscale geoscience datasets are available. is biggest drilling mineral-exploration zone China. This study used three-dimensional (3D) modeling, geology, ore body alteration to extract synthesize mineralization information analyze exploration targeting Zhaoxian deposit northwestern Peninsula. methodology results summarized as follows: regional Jiaojia fault key criterion of...
Due to the rapid development of electronic technologies, it tends be diverse access networks besides personal computers. Thanks is possible for us learn on net anytime and anywhere, which propels progress in e-learning dramatically, some new research issues arise at same time. Firstly, owing differences processing power, storage display resolution terminals, will lead problems that content sent heterogeneous terminals. Secondly, since personalized learning pattern popular modern education,...