A5100453458- Topic Modeling
- Natural Language Processing Techniques
- Atrial Fibrillation Management and Outcomes
- Advanced Text Analysis Techniques
- Multimodal Machine Learning Applications
- Text and Document Classification Technologies
- Advanced Computational Techniques and Applications
- Web Data Mining and Analysis
- Dementia and Cognitive Impairment Research
- Fuzzy Logic and Control Systems
- Cardiac Arrhythmias and Treatments
- Biomedical Text Mining and Ontologies
- Advanced Graph Neural Networks
- Advanced Image and Video Retrieval Techniques
- Sentiment Analysis and Opinion Mining
- Domain Adaptation and Few-Shot Learning
- Machine Learning in Healthcare
- Cancer, Stress, Anesthesia, and Immune Response
- Software Engineering Research
- Schizophrenia research and treatment
- Video Analysis and Summarization
- Mental Health Research Topics
- Diabetes Management and Education
- Cardiac Imaging and Diagnostics
- Technology Adoption and User Behaviour
King's College London
2020-2025
Chengdu University of Traditional Chinese Medicine
2021-2025
Dalian University of Technology
2013-2025
Dalian Municipal Central Hospital
2014-2025
National University of Defense Technology
2013-2024
South London and Maudsley NHS Foundation Trust
2021-2024
Southwestern Medical Center
2024
Sichuan University
2020-2024
Chinese PLA General Hospital
2024
Beijing Jiaotong University
2018-2024
Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Zhongqiang Huang, Fei Kewei Tu. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.
Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Zhongqiang Huang, Fei Kewei Tu. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.
Abstract Motivation: The synapse is integral to the function of brain and may be an important source dysfunction underlying many neuropsychiatric disorders. Consequently, it excellent candidate for large-scale genomic proteomic study. However, while tools databases available annotation high-throughput DNA protein are generally robust, a comprehensive resource dedicated integration information about lacking. Results: We present integrated database, called SynaptomeDB, retrieve annotate genes...
Relation extraction (RE) is a fundamental process in constructing knowledge graphs. However, previous methods on relation suffer sharp performance decline short and noisy social media texts due to lack of contexts. Fortunately, the related visual contents (objects their relations) posts can supplement missing semantics help extract relations precisely. We introduce multimodal (MRE), task that identifies textual with clues. To tackle this problem, we present large-scale dataset which contains...
A three-dimensional (3D) continuous pulse arterial spin labeling (ASL) technique was used to investigate cerebral blood flow (CBF) changes in patients with Alzheimer's disease (AD), amnestic mild cognitive impairment (aMCI), and age- sex-matched healthy controls.Three groups were recruited for comparison, 24 AD patients, 17 MCI 21 control subjects. Three-dimensional ASL scans covering the entire brain acquired a 3.0 T magnetic resonance scanner. Spatial processing performed statistical...
Mild cognitive impairment (MCI) is the prodromal stage of Alzheimer's disease (AD), so identification related risk factors can be helpful. Although association between type 2 diabetes mellitus (T2DM) and these modest changes in cognition well established, whether T2DM will promote transformation MCI into AD not a unified conclusion.This study aims to explore relationship elderly population living community Shanghai, People's Republic China.A total 197 participants were included study. They...
Named Entity Recognition (NER) in social media posts is challenging since texts are usually short and contexts lacking. Most recent works show that visual information can boost the NER performance images provide complementary contextual for texts. However, image-level features ignore mapping relations between fine-grained objects textual entities, which results error detection entities with different types. To better exploit NER, we propose an adversarial gated bilinear attention neural...
Purpose The purpose of this paper is to apply the self-determination theory (SDT) propose a research model that incorporates SDT framework and contextual variables as determinants self-identity social identity mediating constructs predict individuals’ intentions toward donation crowdfunding in China. Design/methodology/approach Structural equation modeling used analyze data collected from Findings results indicate collectively or separately mediate effect exerted by sense self-worth, face...
To update the efficacy and safety of different drugs for treatment patients with hyperphosphatemia in chronic kidney disease, we conducted a network meta-analysis 22 therapies uncontrolled disease (CKD). All randomized controlled trials on published from January 2013 to November 2023 were searched CNKI, VIP database, Wanfang PubMed, Scopus, Cochrane databases. Meta-analysis was used evaluate serum phosphorus, calcium levels, total effective rate adverse events Data collection quality...
Article Abstract Objective: To investigate the relationship between microRNA107 (miRNA107) and BACE1 messenger RNA (mRNA) gene expressions in plasma their diagnostic capability to distinguish subjects with amnestic mild cognitive impairment from healthy controls. Method: We recruited 97 patients Alzheimer's disease according criteria of DSM-IV National Institute Neurologic Communicative Disorders Stroke/Alzheimer's Disease Related Association, 116 impairment, 81 controls January 2012...
The establishment of normative data and screening cut-points for cognitive tasks is important to ensure the effective timely detection mild impairment (MCI) Alzheimer's disease (AD). These need be culturally relevant account known factors that impact on cognition such as age, education, gender. In this study, 1,068 elderly Chinese residents Shanghai completed a comprehensive series part community study with 1027 meeting criteria analysis, age M(SD) = 72.54 (8.40). MCI was detected in 267...
Mobile health apps are increasingly available and used in a clinical context to monitor young people's mood mental health. Despite the benefits of accessibility cost-effectiveness, consumer engagement remains hurdle for uptake continued use. Hundreds mood-monitoring publicly people on app stores; however, few studies have examined perspectives. App store reviews held Google Apple platforms provide large, rich source naturally generated, user reviews. Although commercial developers use these...
In the current study, a novel network model is proposed for text classification based on Graph Attention Networks (GATs) and sentence-transformer embeddings. Most existing methods with pretraining as an input layer still treat words minimum processing unit. However, word embedding not efficient appropriate solution when dealing long texts containing many professional words. This study aims to design capable of handling tasks at multilevel semantic segmentation. The main contribution this...
Abstract Background Atrial cardiopathy is a proposed mechanism of embolic stroke undetermined source (ESUS). Left atrial (LA) strain may identify early prior to structural changes. We aim study the associations between LA strain, ESUS, and fibrillation (AF) detection in ESUS. Methods The population included patients with ESUS noncardioembolic (NCE) presenting Rhode Island Hospital Stroke Center January 2016 June 2017 who underwent transthoracic echocardiography. Speckle tracking...
Semantic event detection is an active research field of video mining in recent years. One the challenging problems how to effectively model temporal and multi-modality characteristics video. In this paper, we employ Conditional Random Fields (CRFs) fuse cues for detection. CRFs are undirected probabilistic models designed segmenting labeling sequence data. Compared with traditional SVM Hidden Markov Models (HMMs), based offers several particular advantages including abilities relax strong...
In GitHub, the pull-based development model enables community contributors to collaborate in a more efficient way. However, distributed and parallel characteristics of this pose potential risk for developers submit duplicate pull-requests (PRs), which increase extra cost project maintenance. To facilitate further studies better understand solve issues introduced by PRs, we construct large dataset historical PRs extracted from 26 popular open source projects GitHub using semi-automatic...
Recent studies have shown that an automated, lifespan-inclusive, transdiagnostic, and clinically based, individualized risk calculator provides a powerful system for supporting the early detection of individuals at-risk psychosis at large scale, by leveraging electronic health records (EHRs). This has been externally validated twice is undergoing feasibility testing clinical implementation. Integration this in routine should be facilitated prospective studies, which are required to address...
Distractor generation task aims to generate incorrect options (i.e., distractors) for multiple choice questions from an article.Existing methods this often utilize a standard encoder-decoder framework. However, these tend semantically similar distractors, since the same article representations are used different distractors. Multiple generated distractors with semantics considered equivalent. Because correct answer is unique, students can eliminate even without reading article. In paper, we...