A5100743423- Advanced Radiotherapy Techniques
- Radiation Therapy and Dosimetry
- Human Pose and Action Recognition
- Advanced Neural Network Applications
- Ocular Oncology and Treatments
- Medical Imaging Techniques and Applications
- Natural Language Processing Techniques
- Video Surveillance and Tracking Methods
- Network Security and Intrusion Detection
- Gait Recognition and Analysis
- Photoacoustic and Ultrasonic Imaging
- Speech Recognition and Synthesis
- Glaucoma and retinal disorders
- Advanced Malware Detection Techniques
- Advanced X-ray and CT Imaging
- Text and Document Classification Technologies
- Advanced Image and Video Retrieval Techniques
- Domain Adaptation and Few-Shot Learning
- Internet Traffic Analysis and Secure E-voting
- Radiation Detection and Scintillator Technologies
- Lung Cancer Treatments and Mutations
- Corneal Surgery and Treatments
- Prostate Cancer Diagnosis and Treatment
- Thermography and Photoacoustic Techniques
- Advanced Numerical Analysis Techniques
Beijing Hospital of Traditional Chinese Medicine
2025
Capital Medical University
2025
Stanford University
2010-2024
Beijing Institute of Petrochemical Technology
2024
Huazhong University of Science and Technology
2024
Central Hospital of Wuhan
2024
Jingning County People's Hospital
2024
First Affiliated Hospital of Dalian Medical University
2023
Jingdong (China)
2020-2023
China Southern Power Grid (China)
2021-2023
Vehicle re-identification (Re-Id) is a challenging task due to the inter-class similarity, intra-class difference, and cross-view misalignment of vehicle parts. Although recent methods achieve great improvement by learning detailed features from keypoints or bounding boxes parts, Re-Id still far being solved. Different existing methods, we propose Parsing-guided Cross-part Reasoning Network, named as PCRNet, for Re-Id. The PCRNet explores parsing learn discriminative part-level features,...
Defining the loss function is an important part of neural network design and critically determines success deep learning modeling. A significant shortcoming conventional functions that they weight all regions in input image volume equally, despite fact system known to be heterogeneous (i.e., some can achieve high prediction performance more easily than others). Here, we introduce a region-specific lift implicit assumption homogeneous weighting for better learning. We divide entire into...
Abstract Traffic classification is widely used in network security and management. Early studies have mainly focused on mapping traffic to different unencrypted applications, but little research has been done of encrypted especially the underlying applications. To address above issues, this paper proposes a encryption model that combines attention mechanisms spatiotemporal features. The firstly uses long short-term memory (LSTM) method analyze continuous flows find temporal correlation...
Malware, such as Trojan Horse, Worms and Spy ware severely threatens Internet. We observed that although malware its variants may vary a lot from content signatures, they share some behavior features at higher level which are more precise in revealing the real intent of malware. This paper investigates technique extraction, presents formal Malware Behavior Feature (MBF) extraction method, proposes malicious feature based detection algorithm. Finally we designed implemented MBF system,...
Elastographic Measurement of the Area and Volume Thermal Lesions Resulting from Radiofrequency Ablation: Pathologic CorrelationTomy Varghese1, Udomchai Techavipoo1 2, Wu Liu1, James A. Zagzebski1, Quan Chen1, Gary Frank1 Fred T. Lee, Jr.3Audio Available | Share
Volumetric modulated arc therapy (VMAT) has recently emerged as a new clinical modality for conformal radiation therapy. The aim of this work is to establish methodology and procedure retrospectively reconstructing the actual dose delivered in VMAT based on pre-treatment cone-beam computed tomography (CBCT) dynamic log files. CBCT was performed before delivery system's files were retrieved after delivery. Actual at control point including MLC leaf positions, gantry angles cumulative monitor...
A novel commercial medical linac system (TrueBeam™, Varian Medical Systems, Palo Alto, CA) allows respiratory-gated volumetric modulated arc therapy (VMAT), a new modality for treating moving tumors with high precision and improved accuracy by allowing regular motion associated patient's breathing during VMAT delivery. The purpose of this work is to adapt previously-developed dose reconstruction technique evaluate the fidelity treatment gated delivery under clinic-relevant periodic related...
Abstract COVID‐19 can lead to adverse outcomes in patients with pre‐existing diseases. Azvudine has been approved for treating China, but the real‐world data is limited. It aimed investigate efficacy of and cardiovascular Patients confirmed diseases are retrospectively enrolled. The primary outcome all‐cause death during hospitalization. Overall, 351 included, a median age 74 years, 44% female. 212 (60.6%) severe cases. used 106 (30.2%) not 245 (69.8%). 72 died After multivariate adjustment,...
Purpose Propagation of contours from high‐quality magnetic resonance (MR) images to treatment planning ultrasound (US) with severe needle artifacts is a challenging task, which can greatly aid the organ contouring in high dose rate (HDR) prostate brachytherapy. In this study, deep learning approach was developed automatize registration procedure for HDR brachytherapy practice. Methods Because lack training labels and difficulty accurate inferior image quality, new segmentation‐based...
Tumor targeting studies using metallic nanoparticles (NPs) have shown that the enhanced permeability and retention effect may not be sufficient to deliver amount of intratumoral intracellular NPs needed for effective in vivo radiosensitization. This work describes a pH-Low Insertion Peptide (pHLIP) targeted theranostic agent enable image-guided NP-enhanced radiotherapy clinically feasible injected NPs. Conventional gadolinium (Gd) were conjugated pHLIPs evaluated vitro radiosensitivity mouse...
There is no consensus for the management of epithelioid trophoblastic tumor (ETT) up to date.ETT rarest form gestational neplasia (GTN). Our goal was assess outcomes and explore prognostic factors patients with ETT through this multicenter retrospective analysis devise a risk-adapted approach clinical management.A total 31 were validated as pathologically between January 2004 June 2021 from three tertiary hospitals. We retrospectively analyzed characteristics, treatments, outcomes,...
Background. Non-small cell lung cancer (NSCLC) is the most prevalent malignant tumor of cancer, for which molecular mechanisms remain unknown. In this study, we identified novel biomarkers associated with pathogenesis NSCLC aiming to provide new diagnostic and therapeutic approaches by bioinformatics analysis. Methods. From Gene Expression Omnibus database, GSE118370 GSE10072 microarray datasets were obtained. Identifying differentially expressed genes (DEGs) between adenocarcinoma normal...
Cross-view gait recognition is a challenging problem when view-interval and pose variation are relatively large. In this paper, we propose Cycle-consistent Attentive Generative Adversarial Networks (CA-GAN) to map different views' images view-consistent photorealistic for cross-view recognition. CA-GAN, the generative network composed of two branches, which simultaneously perceives human's global contexts local body parts information respectively. Moreover, design novel Network (AAN)...
A novel real-time adaptive MV-kV imaging framework for image-guided radiation therapy is developed to reduce the thoracic and abdominal tumor targeting uncertainty caused by respiration-induced intrafraction motion with ultra-low patient dose. In our method, continuous stereoscopic used at beginning of a delivery several seconds measure implanted marker positions. After this period, kV imager switched off except times when no fiducial detected in cine-MV images. The 3D time-varying positions...
To reduce the large computation and storage cost of a deep convolutional neural network, knowledge distillation based methods have pioneered to transfer generalization ability (teacher) network light-weight (student) network. However, these mostly focus on transferring probability distribution softmax layer in teacher thus neglect intermediate representations. In this paper, we propose adversarial better train student Our technique holistically considers both representations distributions...
Text normalization is an important component in text-to-speech system and the difficulty text to disambiguate non-standard words (NSWs). This paper develops a taxonomy of NSWs on basis large scale Chinese corpus, proposes two-stage disambiguation strategy, finite state automata (FSA) for initial classification maximum entropy (ME) classifiers subclass disambiguation. Based above taxonomy, approach achieves F-score 98.53% open test, 5.23% higher than that FSA based approach. Experiments show...