A5067153923- Robotics and Sensor-Based Localization
- Lung Cancer Diagnosis and Treatment
- Gastrointestinal Bleeding Diagnosis and Treatment
- Advanced Vision and Imaging
- Robot Manipulation and Learning
- Microbial Natural Products and Biosynthesis
- Advanced Neural Network Applications
- Soft Robotics and Applications
- Advanced Image and Video Retrieval Techniques
- Genomics and Phylogenetic Studies
- Surgical Simulation and Training
- COVID-19 diagnosis using AI
- AI in cancer detection
- Plant biochemistry and biosynthesis
- Human Pose and Action Recognition
- Pediatric health and respiratory diseases
- Anatomy and Medical Technology
- Head and Neck Surgical Oncology
- Medical Imaging and Analysis
- Fungal Biology and Applications
- Spinal Fractures and Fixation Techniques
- Medical Imaging and Pathology Studies
- Immune Cell Function and Interaction
- Augmented Reality Applications
- Tracheal and airway disorders
Zhejiang University
2023-2024
Guangdong University of Technology
2024
Microbiology Institute of Shaanxi
2022-2024
Institute of Microbiology
2022-2024
Chinese Academy of Sciences
2022-2024
University of Chinese Academy of Sciences
2022-2024
Zhejiang University of Technology
2021-2024
Dalian Medical University
2016
Abstract The unequal distribution of medical resources and scarcity experienced practitioners confine access to bronchoscopy primarily well-equipped hospitals in developed regions, contributing the unavailability bronchoscopic services underdeveloped areas. Here, we present an artificial intelligence (AI) co-pilot bronchoscope robot that empowers novice doctors conduct lung examinations as safely adeptly colleagues. system features a user-friendly, plug-and-play catheter, devised for...
Bacterial small molecule metabolites such as adenosine-diphosphate-d-
// Jing Kong 1 , Yong Luo 2 Dong Jin Fan An Wenyuan Zhang Lilu Liu Jiao Li Shimeng Fang Xiaojie Xuesong Yang 3 Bingcheng Lin 2, 4 Tingjiao College of Stomatology, Dalian Medical University, Dalian, China Faculty Chemical, Environmental and Biological Science Technology, Technology Department Biochemistry Molecular Biology, Liaoning Provincial Core Lab Glycobiology Glycoengineering, Biotechnology, Institute Chemical Physics, Chinese Academy Sciences, Correspondence to: Liu, email:...
Abstract Modular polyketide synthases (PKSs) are capable of synthesizing diverse natural products with fascinating bioactivities. Canonical enoyl‐CoA hydratases (ECHs) components the β ‐branching cassette that modifies chain by adding a ‐methyl branch. Herein, it is demonstrated deletion an atypical ECH Q domain (featuring 280 residue) Art21, didomain protein contains and thioesterase (TE) domain, reprograms assembly line from tetracyclic aurantinins (ARTs) to bicyclic auritriacids (ATAs)...
Common medical image segmentation tasks require large training datasets with pixel-level annotations which are very expensive and time-consuming to prepare. To overcome such limitation achieve the desired accuracy, a novel Weakly-Interactive-Mixed Learning (WIML) framework is proposed by efficiently using weak labels. On one hand, utilize labels reduce annotation time for high-quality strong designing Weakly-Interactive Annotation (WIA) part of WIML prudently introduces interactive learning...
Real-time visual-aided navigation and path strategy for pneumonoconiosis efficient 3D visualization of pulmonary vessels are great research clinical significance in the treatment lung diseases. The complex structure tissue limits application deep learning vascular due to lack labeling datasets. Also, existing methods have large computational complexity low efficiency. This study proposes a method high-quality based on low-cost segmentation fast reconstruction, consisting three steps: 1)...
Lung disease has become a leading cause of disease‐related death, making it one the most serious health problems in world. Due to advantages fast postoperative recovery and small trauma, transbronchoscopic biopsy main method for diagnosis lung diseases. However, limited by large outer diameter bronchoscope, insufficient flexibility needle, lack force sensing, traditional operation is simple rough, difficult enter deep narrow areas examination. In this article, flexible robot with...
Intra-operative target pose estimation is fundamental in minimally invasive surgery (MIS) to guiding surgical robots. This task can be fulfilled by the 2-D/3-D rigid registration, which aligns anatomical structures between intra-operative 2-D fluoroscopy and pre-operative 3-D computed tomography (CT) with annotated information. Although this technique has been researched for decades, it still challenging achieve accuracy, robustness efficiency simultaneously. In paper, a novel...
Visual localization for planar moving robot is important to various indoor service robotic applications. To handle the textureless areas and frequent human activities in environments, a novel robust visual algorithm which leverages dense correspondence sparse depth proposed. The key component minimal solution computes absolute camera pose with one 3D-2D 2D-2D correspondence. advantages are obvious two aspects. First, robustness enhanced as sample set estimation maximal by utilizing all...
The stereochemistry of aurantinin was determined by spectroscopic and computational analysis with the assistance biosynthetic studies. latter method provided critical evidence for assignment configuration 3-ketosugar moiety.
Current RGB-based 6D object pose estimation methods have achieved noticeable performance on datasets and real world applications. However, predicting from single 2D image features is susceptible to disturbance changing of environment textureless or resemblant surfaces. Hence, generally achieve less competitive results than RGBD-based methods, which deploy both 3D structure features. To narrow down this gap, paper proposes a framework for that learns implicit information 2 RGB images....
In order to accurately identify large-scale <strong>sugarcane</strong> plantations, pre-processed Sentinel-2A <strong>remote sensing imagery</strong> data from three different time periods were used in this study. The vegetation index features NDVI, BI2 and S2REP calculated the optimal texture feature, homogeneity, was selected using principal component analysis grey level co-occurrence matrix. A random forest algorithm then construct identification models with combinations of features,...
Visual localization for planar moving robot is important to various indoor service robotic applications. To handle the textureless areas and frequent human activities in environments, a novel robust visual algorithm which leverages dense correspondence sparse depth proposed. The key component minimal solution computes absolute camera pose with one 3D-2D 2D-2D correspondence. advantages are obvious two aspects. First, robustness enhanced as sample set estimation maximal by utilizing all...
Current RGB-based 6D object pose estimation methods have achieved noticeable performance on datasets and real world applications. However, predicting from single 2D image features is susceptible to disturbance changing of environment textureless or resemblant surfaces. Hence, generally achieve less competitive results than RGBD-based methods, which deploy both 3D structure features. To narrow down this gap, paper proposes a framework for that learns implicit information 2 RGB images....
Current monocular-based 6D object pose estimation methods generally achieve less competitive results than RGBD-based methods, mostly due to the lack of 3D information. To make up this gap, paper proposes a geometric volume based method with short baseline two-view setting. By constructing in space, we combine features from two adjacent images same space. Then network is trained learn distribution position keypoints volume, and robust soft RANSAC solver deployed solve closed form. balance...