A5103153719- Image Enhancement Techniques
- MRI in cancer diagnosis
- Video Surveillance and Tracking Methods
- Radiomics and Machine Learning in Medical Imaging
- Advanced Image Processing Techniques
- Visual Attention and Saliency Detection
- Advanced Vision and Imaging
- Advanced MRI Techniques and Applications
- Head and Neck Cancer Studies
- Fire Detection and Safety Systems
- Advanced Image Fusion Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- Computer Graphics and Visualization Techniques
- Advanced Neuroimaging Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- Adversarial Robustness in Machine Learning
- Digital Media Forensic Detection
- Image and Signal Denoising Methods
- VLSI and Analog Circuit Testing
- Bone and Joint Diseases
- Integrated Circuits and Semiconductor Failure Analysis
- 3D Shape Modeling and Analysis
- Advanced Neural Network Applications
- UAV Applications and Optimization
Chinese Academy of Medical Sciences & Peking Union Medical College
2023-2025
University of Electronic Science and Technology of China
2020-2024
Wuhan University
2023
Northeastern University
2021
IVIM is a useful quantitative tool for predicting prognosis, but it labor-intensive. Simplified b-value settings and post-processing could be more practicable clinical applications. To assess the value of model-free parameters (virtual MR elastography [vMRE] signature-index [S-index]) derived from in evaluating pathological indicators long-term survival nasopharyngeal carcinoma (NPC), to compare those with model-based parameters. Prospective. One hundred six patients (median: 49.5 years,...
Due to the flexible training requirement and appealing generalization ability, unpaired image dehazing has received increasing attention in coping with real-world hazy images. However, most of existing methods rely on loose dehazing-hazing cycle constraint, which makes it hard eliminate poor-quality results when using a powerful hazing network process. To address this issue, paper proposes simple yet efficient Adversarial Deformation Constraint (ADC). More specifically, we sequentially...
Abstract Background To investigate the potential of synthetic MRI (SyMRI) in prognostic assessment patients with nonmetastatic nasopharyngeal carcinoma (NPC), and predictive value when combined diffusion-weighted imaging (DWI) as well clinical factors. Methods Fifty-three NPC who underwent SyMRI were prospectively included. 10th Percentile, Mean, Kurtosis, Skewness T1, T2, PD maps ADC obtained from primary tumor. Cox regression analysis was used for analyzing association between DWI...
Due to the lack of natural scene and haze prior information, it is greatly challenging completely remove from a single image without distorting its visual content. Fortunately, real-world usually presents non-homogeneous distribution, which provides us with many valuable clues in partial well-preserved regions. In this paper, we propose Non-Homogeneous Haze Removal Network (NHRN) via artificial bidimensional graph reasoning. Firstly, employ gamma correction iteratively simulate multiple...
This article introduces an on-chip anomaly monitoring system design approach that is based on thermal profiling and side-channel analysis. The strategy aims at the realization of nonintrusive hardware Trojan (HT) detection over lifetime circuit under test (CUT). To evaluate capability proposed HT system, electrothermal coupling modeled as part simulation technique, which associates local activities with circuit-level power consumption using a standard electrical simulator. monitor profiles...
Efficient object detection and tracking from remote sensing video data acquired by unmanned aerial vehicles (UAVs) has significant implications in various domains, such as scene understanding, traffic surveillance, military operations. Although the modern transformer-based trackers have demonstrated superior accuracy, they often require extensive training time to achieve convergence, information templates is not fully utilized integrated into tracking. To accelerate convergence further...
The main challenge for single image dehazing is the lack of effective prior information restoration. To address this issue, in paper, we propose to generate artificial multiple shots simulating images captured under different haze degrees, and two context reasoning modules are developed describe relationship across spatial regions shots. It brings benefits inhomogeneous distribution. First, within one shot, occluded location could be recovered with help other clear regions, which share...
Image understanding under the foggy scene is greatly challenging due to inhomogeneous visibility deterioration. Although various image dehazing methods have been proposed, they usually aim improve (such as, PSNR/SSIM) in pixel space rather than feature space, which critical for perception of computer vision. Due this mismatch, existing are limited or even adverse facilitating understanding. In paper, we propose a generalized deep refinement module minimize difference between clear images and...
Despite demonstrating superior rate-distortion (RD) performance, learning-based image compression (LIC) algorithms have been found to be vulnerable malicious perturbations in recent studies. Adversarial samples these studies are designed attack only one dimension of either bitrate or distortion, targeting a submodel with specific ratio. However, adversaries real-world scenarios neither confined singular dimensional attacks nor always control over ratios. This variability highlights the...
Recent diffusion-based Single-image 3D portrait generation methods typically employ 2D diffusion models to provide multi-view knowledge, which is then distilled into representations. However, these usually struggle produce high-fidelity models, frequently yielding excessively blurred textures. We attribute this issue the insufficient consideration of cross-view consistency during process, resulting in significant disparities between different views and ultimately leading In paper, we address...