A5101434323- Topic Modeling
- Natural Language Processing Techniques
- Advanced Graph Neural Networks
- Multimodal Machine Learning Applications
- Surgical Simulation and Training
- Domain Adaptation and Few-Shot Learning
- Colorectal Cancer Screening and Detection
- Advanced Text Analysis Techniques
- Anatomy and Medical Technology
- Semantic Web and Ontologies
- Data Quality and Management
- Head and Neck Surgical Oncology
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Lung Cancer Diagnosis and Treatment
- Gastrointestinal Bleeding Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Advanced Data Compression Techniques
- Color perception and design
- Artificial Intelligence in Healthcare and Education
- Cardiac, Anesthesia and Surgical Outcomes
- Aesthetic Perception and Analysis
- Graph Theory and Algorithms
- Brain Tumor Detection and Classification
- Image Enhancement Techniques
Xuzhou Medical College
2024-2025
Inner Mongolia University of Science and Technology
2025
Chinese University of Hong Kong
2023-2024
Institute of Computing Technology
2020-2024
Chinese Academy of Sciences
2020-2024
University of Chinese Academy of Sciences
2020-2024
Wenzhou University
2024
Shanghai University
2024
New York University Shanghai
2024
Shijiazhuang University
2024
Zixuan Li, Saiping Guan, Xiaolong Jin, Weihua Peng, Yajuan Lyu, Yong Zhu, Long Bai, Wei Jiafeng Guo, Xueqi Cheng. Proceedings of the 60th Annual Meeting Association for Computational Linguistics (Volume 2: Short Papers). 2022.
Besides entity-centric knowledge, usually organized as Knowledge Graph (KG), events are also an essential kind of knowledge in the world, which trigger spring up event-centric representation form like Event KG (EKG). It plays increasingly important role many downstream applications, such search, question-answering, recommendation, financial quantitative investments, and text generation. This paper provides a comprehensive survey EKG from history, ontology, instance, application views....
Knowledge Graph Question Answering aims to automatically answer natural language questions via well-structured relation information between entities stored in knowledge graphs. When faced with a complex question compositional semantics, query graph generation is practical semantic parsing-based method. But existing works rely on heuristic rules limited coverage, making them impractical more questions. This paper proposes Director-Actor-Critic framework overcome these challenges. Through...
Despite the availability of computer-aided simulators and recorded videos surgical procedures, junior residents still heavily rely on experts to answer their queries. However, expert surgeons are often overloaded with clinical academic workloads limit time in answering. For this purpose, we develop a question-answering system facilitate robot-assisted scene activity understanding from videos. Most existing visual question answering (VQA) methods require an object detector regions based...
Researchers have long converged that the evolution of a Social Networking Service (SNS) platform is driven by interplay between users' preferences (reflected in user-item consumption behavior) and social network structure user-user interaction behavior), with both kinds behaviors change from time to time. However, traditional approaches either modeled these two an isolated way or relied on static assumption SNS. Thus, it still unclear how do roles historical dynamic affect SNSs. Furthermore,...
Surgical tool segmentation and action recognition are fundamental building blocks in many computer-assisted intervention applications, ranging from surgical skills assessment to decision support systems. Nowadays, learning-based approaches outperform classical methods, relying, however, on large, annotated datasets. Furthermore, algorithms often trained make predictions isolation each other, without exploiting potential cross-task relationships. With the EndoVis 2022 SAR-RARP50 challenge, we...
Objective There have been proposals that vitamin D may be associated with a reduction in the incidence of anxiety disorders. However, findings thus far inconsistent, warranting further investigation. The purpose this paper is to explore link between serum and anxiety. Methods Data are from National Health Nutrition Examination Survey (NHANES) United States 2007 2012. Study included total 12,232 participants, through multivariate logistic regression study relationship anxiety, smooth curve...
In recent years, purchasing medications online has become increasingly popular. However, occasional quality issues have arisen with drugs bought online. As a result, we need effective monitoring of medicines sold To address this issues, several countries begun to implement the Marketing Authorization Holder (MAH) system enhance drug control. Consequently, paper develops four-party evolutionary game model that includes government, holder, agent seller and third-party platform in...
A Temporal Knowledge Graph (TKG) is a sequence of KGs with respective timestamps, which adopts quadruples in the form (subject, relation, object, timestamp) to describe dynamic facts. TKG reasoning has facilitated many real-world applications via answering such queries as (query entity, query ?, future about future. This actually matching task between and candidate entities based on their historical structures, reflect behavioral trends at different timestamps. In addition, recent provide...
Abstract Objective The present study aimed to evaluate the risk factors for gestational diabetes mellitus (GDM) and build validate an early prediction model of GDM by comparing differences in indicators first trimester pregnancy between pregnant women with non-gestational (NGDM). Thus, this provided a theoretical basis intervention GDM. Methods A total 6000 who underwent routine prenatal examination Qinhuangdao Maternal Child Health Hospital (Qinhuangdao City, Hebei Province, China) from...
Large language models (LLMs) have recently demonstrated remarkable performance across various Natual Language Processing tasks. In the field of multi-hop reasoning, Chain-of-thought (CoT) prompt method has emerged as a paradigm, using curated stepwise reasoning demonstrations to enhance LLM's ability reason and produce coherent rational pathways. To ensure accuracy, reliability, traceability generated answers, many studies incorporated information retrieval (IR) provide LLMs with external...
Recent advancements in Surgical Visual Question Answering (Surgical-VQA) and related region grounding have shown great promise for robotic medical applications, addressing the critical need automated methods personalized surgical mentorship. However, existing models primarily provide simple structured answers struggle with complex scenarios due to their limited capability recognizing long-range dependencies aligning multimodal information. In this paper, we introduce Surgical-LVLM, a novel...
Scripts are structured sequences of events together with the participants, which extracted from texts. Script event prediction aims to predict subsequent given historical in script. Two kinds information facilitate this task, namely, event-level and script-level information. At level, existing studies view an as a verb its while neglecting other useful properties, such state participants. script most only consider single sequence corresponding one common protagonist. In paper, we propose...
Legal case retrieval, which aims to find relevant cases based on a short description, serves as an important part of modern legal systems. Despite the success existing retrieval methods Pretrained Language Models, there are still two issues in that have not been well considered before. First, underestimate semantics associations among elements, e.g., law articles and crimes, played essential role retrieval. These only adopt pre-training language model encode whole case, instead...