A5016388137- Machine Learning in Healthcare
- Topic Modeling
- Advanced MRI Techniques and Applications
- Biomedical Text Mining and Ontologies
- Medical Image Segmentation Techniques
- Medical Imaging and Analysis
- Artificial Intelligence in Healthcare
- Information Retrieval and Search Behavior
- Natural Language Processing Techniques
- Functional Brain Connectivity Studies
- Advanced Neuroimaging Techniques and Applications
- Medical Imaging Techniques and Applications
- Recommender Systems and Techniques
- COVID-19 diagnosis using AI
- MRI in cancer diagnosis
- Power System Reliability and Maintenance
- Spine and Intervertebral Disc Pathology
- Semantic Web and Ontologies
- Liver Disease Diagnosis and Treatment
- Advanced Text Analysis Techniques
- COVID-19 Clinical Research Studies
- Asthma and respiratory diseases
- Dementia and Cognitive Impairment Research
- Biometric Identification and Security
- Brain Tumor Detection and Classification
Southern Medical University
2016-2025
Jinling Institute of Technology
2025
Zhejiang Chinese Medical University
2022-2025
Second Affiliated Hospital of Nanjing Medical University
2025
Chinese Academy of Fishery Sciences
2025
The University of Texas Health Science Center at Houston
2018-2024
Museum of Fine Arts, Houston
2024
Lanzhou University
2024
Yale University
2024
Renmin University of China
2022-2024
Content-based image retrieval (CBIR) techniques have currently gained increasing popularity in the medical field because they can use numerous and valuable archived images to support clinical decisions. In this paper, we concentrate on developing a CBIR system for retrieving brain tumors T1-weighted contrast-enhanced MRI images. Specifically, when user roughly outlines tumor region of query image, database same pathological type are expected be returned. We propose novel feature extraction...
Abstract Importance The study highlights the potential of large language models, specifically GPT-3.5 and GPT-4, in processing complex clinical data extracting meaningful information with minimal training data. By developing refining prompt-based strategies, we can significantly enhance models’ performance, making them viable tools for NER tasks possibly reducing reliance on extensive annotated datasets. Objectives This quantifies capabilities GPT-4 named entity recognition (NER) proposes...
Objective: This study quantifies the capabilities of GPT-3.5 and GPT-4 for clinical named entity recognition (NER) tasks proposes task-specific prompts to improve their performance. Materials Methods: We evaluated these models on two NER tasks: (1) extract medical problems, treatments, tests from notes in MTSamples corpus, following 2010 i2b2 concept extraction shared task, (2) identifying nervous system disorder-related adverse events safety reports vaccine event reporting (VAERS). To GPT...
Spine parsing (i.e., multi-class segmentation of vertebrae and intervertebral discs (IVDs)) for volumetric magnetic resonance (MR) image plays a significant role in various spinal disease diagnoses treatments spine disorders, yet is still challenge due to the inter-class similarity intra-class variation images. Existing fully convolutional network based methods failed explicitly exploit dependencies between different structures. In this article, we propose novel two-stage framework named...
The COVID-19 pandemic swept across the world rapidly, infecting millions of people. An efficient tool that can accurately recognize important clinical concepts from free text in electronic health records (EHRs) will be valuable to accelerate research. To this end, study aims at adapting existing CLAMP natural language processing quickly build SignSym, which extract signs/symptoms and their 8 attributes (body location, severity, temporal expression, subject, condition, uncertainty, negation,...
Predicting outcomes of patients with COVID-19 at an early stage is crucial for optimised clinical care and resource management, especially during a pandemic. Although multiple machine learning models have been proposed to address this issue, because their requirements extensive data preprocessing feature engineering, they not validated or implemented outside original study site. Therefore, we aimed develop accurate transferrable predictive on hospital admission COVID-19.In study, developed...
In recent years, the digital economy has shown great potential in regard to driving social production and development. context of construction villages, deep integration agricultural development injected new vitality into improving quality efficiency production, becoming an important way promote sustainable Based on panel data 31 provinces China from 2012 2021, study utilizes entropy method measure level high-quality agriculture. Additionally, this explores impact mechanism agriculture by...
Large reasoning models (LRMs) like OpenAI-o1 have demonstrated impressive long stepwise capabilities through large-scale reinforcement learning. However, their extended processes often suffer from knowledge insufficiency, leading to frequent uncertainties and potential errors. To address this limitation, we introduce \textbf{Search-o1}, a framework that enhances LRMs with an agentic retrieval-augmented generation (RAG) mechanism Reason-in-Documents module for refining retrieved documents....
Background MRI is the most commonly used imaging method for diagnosing anterior cruciate ligament (ACL) injuries. However, interpretation of knee time‐intensive and depends on clinical experience reader. An automated detection system based a deep‐learning algorithm may improve time reliability. Purpose To determine feasibility using deep learning approach to detect ACL injuries within joint MRI. Study Type Retrospective. Population In all, 163 subjects with an tear 245 intact ACL. There were...
a 五邑大学生物科技与大健康学院 广东江门 529020) ( b 深圳大学生命与海洋科学学院 广东深圳 518060) c 兰州大学药学院 兰州 730000) 摘要 合成了系列 2β-acetoxyferruginol 去醋酸基骨架衍生物(1～24), 并测定了其 α-葡萄糖苷酶抑制活性.结果表明: 化合物 1～24 均有较好的 α-葡萄糖苷酶抑制作用.其中(3R,4aS,10aS)-6-羟基-1,1,4-三甲基-1,2,3,4,4a,9,10,10-八氢邻蒽-3-基-4-(三氟甲基)苯甲酸酯(15)抑制 α-葡萄糖苷酶活性最强[IC50＝(23.91±2.34)μmol/L], 是阿卡波糖抑制活性的 23.6 倍. 构效关系分析表明三氟甲基的引入更有利于提高化合物的活性.动力学结果显示化合物 15 为可逆非竞争性的 α-葡 萄糖苷酶抑制剂.3D 荧光结果表明化合物 与 α-葡萄糖苷酶的结合可改变 α-葡萄糖苷酶的构象.分子对接结果显示化
The consistent and persuasive evidence illustrating the influence of social determinants on health has prompted a growing realization throughout care sector that enhancing equity will likely depend, at least to some extent, addressing detrimental determinants. However, detailed (SDoH) information is often buried within clinical narrative text in electronic records (EHRs), necessitating natural language processing (NLP) methods automatically extract these details. Most current NLP efforts for...
Background Asthma exacerbation is an acute or subacute episode of progressive worsening asthma symptoms and can have a significant impact on patients’ quality life. However, efficient methods that help identify personalized risk factors make early predictions are lacking. Objective This study aims to use advanced deep learning models better predict the exacerbations explore potential involved in asthma. Methods We proposed novel time-sensitive, attentive neural network using clinical...
: Developing clinical natural language processing systems often requires access to many documents, which are not widely available the public due privacy and security concerns. To address this challenge, we propose develop methods generate synthetic notes evaluate their utility in real tasks.: We implemented 4 state-of-the-art text generation models, namely CharRNN, SegGAN, GPT-2, CTRL, for History Present Illness section. then manually annotated entities randomly selected 500 generated from...
Summary Background Previous studies have demonstrated an association between nonselective beta‐blockers (NSBBs) and lower risk of hepatocellular carcinoma (HCC) in cirrhosis. However, there has been no population‐based study investigating the HCC among cirrhotic patients treated using carvedilol. Aims To determine with NSBBs including Methods This retrospective cohort utilised Cerner Health Facts database United States from 2000 to 2017. Kaplan‐Meier estimate, Cox proportional hazards...
We aimed to develop a distributed, immutable, and highly available cross-cloud blockchain system facilitate federated data analysis activities among multiple institutions.
A hitherto unreported catalyst-free ring expansion reaction of tetrahydroisoquinolines with o -alkynylarylaldehydes for the construction dibenzo[ b , d ]azepine skeleton is described.
Personal recognition using palm–vein patterns has emerged as a promising alternative for human because of its uniqueness, stability, live body identification, flexibility, and difficulty to cheat. With the expanding application pattern recognition, corresponding growth database resulted in long response time. To shorten time this paper proposes simple useful classification identification based on principal direction features. In registration process, Gaussian-Radon transform is adopted...
The rapid evolution of artificial intelligence (AI) in conjunction with recent updates dual antiplatelet therapy (DAPT) management guidelines emphasizes the necessity for innovative models to predict ischemic or bleeding events after drug-eluting stent implantation. Leveraging AI dynamic prediction has potential revolutionize risk stratification and provide personalized decision support DAPT management.
Learning distributional representation of clinical concepts (e.g., diseases, drugs, and labs) is an important research area deep learning in the medical domain. However, many existing relevant methods do not consider temporal dependencies along longitudinal sequence a patient's records, which may lead to incorrect selection contexts. To address this issue, we extended three popular concept embedding methods: word2vec, positive pointwise mutual information (PPMI) FastText, time-sensitive...