- Cryptography and Data Security
- Privacy-Preserving Technologies in Data
- Quantum Computing Algorithms and Architecture
- Quantum Mechanics and Applications
- Quantum Information and Cryptography
- Brain Tumor Detection and Classification
- Internet Traffic Analysis and Secure E-voting
- COVID-19 diagnosis using AI
- User Authentication and Security Systems
- Complexity and Algorithms in Graphs
- Advanced Malware Detection Techniques
- Chaos-based Image/Signal Encryption
- Human Pose and Action Recognition
- AI in cancer detection
- Deception detection and forensic psychology
- Digital Mental Health Interventions
- Radiomics and Machine Learning in Medical Imaging
- Parathyroid Disorders and Treatments
- Advanced Neural Network Applications
- Cloud Data Security Solutions
- Gait Recognition and Analysis
- Medical Image Segmentation Techniques
- Recommender Systems and Techniques
- Image Retrieval and Classification Techniques
- Anomaly Detection Techniques and Applications
University of Electronic Science and Technology of China
2014-2023
With the rapid development of Mobile Internet and Industrial Things, a variety applications put forward an urgent demand for user device identity recognition. Digital with hidden characteristics is essential both individual users physical devices. assistance multimodalities as well fusion strategies, recognition can be more reliable robust. In this survey, we turn to investigate concepts limitations unimodal recognition, motivation, advantages multimodal summarize technologies via feature...
Pulmonary tuberculosis (TB), the most prevalent form of TB, remains a major global public health concern, contributing to more than million deaths each year. The accurate and timely diagnosis this disease is paramount importance for effective control treatment. Chest X-ray (CXR) images have emerged as valuable tool screening lung diseases, including owing their cost-effectiveness non-invasiveness. Despite advancements in technology, challenges associated with interpreting CXR persist,...
Personal health record (PHR) is a patient-centric model of information exchange, which greatly facilitates the storage, access, and share personal information. In order to valuable resources reduce operational cost, PHR service providers would like store applications data in cloud. The private may be exposed unauthorized organizations or individuals since patient lost physical control their Ciphertext-policy attribute-based signcryption promising solution design cloud-assisted secure sharing...
Histopathology diagnosis is an important standard for breast tumors identifying. However, histopathology image analysis complex, tedious and error-prone, due to the super-resolution image. In recent years, deep learning technology has been successfully applied made great progress. The well-known neural networks usually have tens of million parameters, which consume much memory deploy state-of-the-art model. addition, rely on high-performance hardware resources, impede deployment model...
Domain adaptation methods train a model to find similar feature representations between source and target domain. Recent leverage self-supervised learning discover the analogous of two domains. However, prior have three significant drawbacks: (1) leveraging pretext tasks that are susceptible low-level representations, (2) aligning domains using adversarial loss without considering if extracted features (3) models not flexible accommodate various proportions labels, i.e., they assume labels...
Abstract With the rapid development of Internet Vehicles, a great amount vehicles are connected to Vehicles environment. These can exchange data each other and upload cloud server for backup. Since contains some sensitive information, it is necessary encrypt before uploading server. However, this method severely hampers utilization because tremendous difficulty in search encrypted data. For addressing above issue, we propose novel certificateless public key encryption supporting equality...
PHR System is a favorable platform for personal health information exchange. In order to ensure that the not falsified and leaked by malicious users, we use attribute-based signcryption technology provide secure reliable data protection. At same time, in prevent users from accessing system collusion of attributes, proposes revocable cloud-assisted scheme which using broadcast encryption key segmentation realize user revocation function. Moreover, proposed proven be confidentiality...
In the area of medical image segmentation, spatial information can be further used to enhance segmentation performance. And 3D convolution is mainly better utilize information. However, how in 2D still a challenging task. this paper, we propose an network based on reinforcement learning (RLSegNet), which translate process into serial decision-making problem. The proposed RLSegNet U-shaped network, composed three components: feature extraction Mask Prediction Network (MPNet), and up-sampling...
In this paper, we propose a scheme of quantum information splitting arbitrary three-qubit state by using seven-qubit entangled as channel. The sender Alice first performs Bell-state measurements (BSMs) on her qubits pairs respectively and tells measurement outcome to authorizers Bob reconstruct the original state, then Charlie should carries out single-qubit (SQM) his qubits. According results from Charlie, can applying an appropriate unitary operation. After analyzing, method achieved...
Intelligent security expects to avoid the occurrence of robbery, theft, and other undesirable situations through video surveillance. In surveillance, images human faces undimmed are not easily available, so pedestrian re-identification (person ReID) is an alternative technique which attracts a mount researchers attention. Person ReID used match across cameras. Due interference shooting angle camera quality, it difficult obtain high resolution, no obstructions, simple backgrounds similar...
Human skeleton data, which has served in the aspect of human activity recognition, ought to be most representative biometric characteristics due its intuitivity and visuality. The state-of-the-art approaches mainly focus on improving modeling spatial correlations within graph topologies. However, interframes motional representations are also vital importance, we argue that they worth paying attention exploring. Therefore, a temporal refinement module with contrastive learning mechanism is...
Content-based (<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M1"><mml:mrow><mml:mtext>CB</mml:mtext></mml:mrow></mml:math>) and collaborative filtering id="M2"><mml:mrow><mml:mtext>CF</mml:mtext></mml:mrow></mml:math>) recommendation algorithms are widely used in modern e-commerce recommender systems id="M3"><mml:mrow><mml:mtext>RSs</mml:mtext></mml:mrow></mml:math>) to improve user experience of personalized services. Item content features user-item rating data primarily...
Most of the existing multi-workflow scheduling algorithms combine into one workflow, and then use a single workflow method to schedule synthetic workflow. This approach merging requires first wait for behind, which reduces response time user experience. To address this gap, paper considers arrival different workflows two conflicting objectives, proposes multi-objective genetic algorithm scheduling. divides each tasks, uses find optimal mapping tasks virtual machines based on task execution...