A5057844830- Machine Learning in Bioinformatics
- Metabolomics and Mass Spectrometry Studies
- Metallurgy and Material Forming
- Acoustic Wave Phenomena Research
- Image Enhancement Techniques
- Metal Forming Simulation Techniques
- COVID-19 diagnosis using AI
- Traditional Chinese Medicine Studies
- Powder Metallurgy Techniques and Materials
- Tribology and Lubrication Engineering
- Advanced Vision and Imaging
- Image and Signal Denoising Methods
- Lung Cancer Diagnosis and Treatment
- Generative Adversarial Networks and Image Synthesis
- Gear and Bearing Dynamics Analysis
- Cell Image Analysis Techniques
- Remote Sensing and LiDAR Applications
- Face recognition and analysis
- Music and Audio Processing
- Intracerebral and Subarachnoid Hemorrhage Research
- Vehicle Noise and Vibration Control
- Digital Imaging for Blood Diseases
- Radiology practices and education
- Structural Engineering and Vibration Analysis
- Video Analysis and Summarization
Winning Health Technology Group (China)
2019-2021
University of Macau
2021
North China University of Technology
2017-2018
Guangxi University
2014-2015
Dantec Dynamics (United Kingdom)
2012
Acute Intracranial hemorrhage (ICH) is a life-threatening disease that requires emergency medical attention, which routinely diagnosed using non-contrast head CT imaging. The diagnostic accuracy of acute ICH on varies greatly among radiologists due to the difficulty interpreting subtle findings and time pressure associated with ever-increasing workload. use artificial intelligence technology may help automate process assist for more prompt better decision-making. In this work, we design deep...
Pinpointing subcellular protein localizations from microscopy images is easy to the trained eye, but challenging automate. Based on Human Protein Atlas image collection, we held a competition identify deep learning solutions solve this task. Challenges included training highly imbalanced classes and predicting multiple labels per image. Over 3 months, 2,172 teams participated. Despite convergence popular networks techniques, there was considerable variety among solutions. Participants...
Image harmonization is a critical task in computer vision, which aims to adjust the foreground make it compatible with background. Recent works mainly focus on using global transformations (i.e., normalization and color curve rendering) achieve visual consistency. However, these models ignore local consistency their huge model sizes limit ability edge devices. In this paper, we propose hierarchical dynamic network (HDNet) adapt features from view for better feature transformation efficient...
Pneumothorax is common but a life-threatening thoracic disease, which difficult to diagnose based on chest X-ray images due its subtle characteristics and low contrast of the disease regions. We aim develop fully automatic pneumothorax segmentation method assist radiologists for timely accurate diagnosis pneumothorax. propose two-stage deep learning method. In first stage, image classified as having or not using an ensemble multiple modified U-Net models. Each model trained multitask...
Abstract The tunable manipulation of sound propagation for on-chip waveguide is crucial acoustic integrated circuits. Here, a non-uniform Mie resonator with anisotropic spatial dispersion characteristic proposed as an adjustable unit to realize transmission across multiple paths within the network metastructure. structure has four waveguides different refractive indexes obtained by zigzag waveguides, which can excite fixed and rotational hyperbolic characteristics simultaneously. On this...
The development of text-to-image generative models has enabled the creation images so realistic that distinguishing between AI-generated and real photos is becoming a challenge. This progress offers new possibilities but also raises concerns over privacy, authenticity, security. Detecting crucial to prevent misuse. To assess generalizability robustness image detection, we present large-scale dataset, referred as WildFake. dataset features cutting-edge generators, wide variety generator...
Artificial intelligence can assist in interpreting chest X-ray radiography (CXR) data, but large datasets require efficient image annotation. The purpose of this study is to extract CXR labels from diagnostic reports based on natural language processing, train convolutional neural networks (CNNs), and evaluate the classification performance CNN using data multiple centers.We collected images corresponding radiology 74,082 subjects as training dataset. linguistic entities relationships...
Recently, video generation techniques have advanced rapidly. Given the popularity of content on social media platforms, these models intensify concerns about spread fake information. Therefore, there is a growing demand for detectors capable distinguishing between AI-generated videos and mitigating potential harm caused by However, lack large-scale datasets from most generators poses barrier to development such detectors. To address this gap, we introduce first detection dataset, GenVideo....
This study reveals a cutting-edge re-balanced contrastive learning strategy aimed at strengthening face anti-spoofing capabilities within facial recognition systems, with focus on countering the challenges posed by printed photos, and highly realistic silicone or latex masks. Leveraging HySpeFAS dataset, which benefits from Snapshot Spectral Imaging technology to provide hyperspectral images, our approach harmonizes class-level data resampling an innovative real-face oriented reweighting...
Statistical energy analysis (SEA) method is an adequate tool to solve complex problems building acoustics. This research on the application in variety of materials as subsystem SEA model performed. For purpose explore relationship between element and its mode, these commonly used are selected determine this relationship. It indicated that properties material have obvious effect modal density overlap members. As consequence, a useful technique account for member be appropriate (statistical...
Coupling Loss Factor (CLF) is a parameter describing building sound loss, which can be stand for energy loss in the process of crossing structure. A low value CLF refers to high insulation performance member. Therefore, reducing coupling favorable way improve insulation. For purpose exploring relationship between properties materials and CLF, commonly used are selected analyze. It indicated that material have obvious effects on CLF. As consequence, some predictions analysis carried out this paper.
In this paper, a theoretical model to evaluate impact sound transmission through homogeneous wall is proposed. The which based on the Statistical Energy Analysis framework exhibits system with room-wall-room. For purpose explore mechanism of wall, reduction index between two rooms are predicted. Meanwhile, variation walls properties also taken into account. results reveal that density, elastic modulus and thickness have diverse effects its insulation can be chosen adequately achieve ideal...
Objective To detect the feasibility and efficiency of bone age(BA) artificial intelligence(AI) estimation based on deep learning features from traditional regions interest(ROI) in hand digital radiographs(DR). Methods BA dataset left DR with 11 858 subjects aged 0 to 18 years Children′s Hospital Shanghai were split training(80.0%) validation (20.0%) set this study. An improved regression convolutional neural networks extreme gradient boosting decision tree method utilized for BA...