A5015082268- Emergency and Acute Care Studies
- Machine Learning in Healthcare
- Electronic Health Records Systems
- Scientific Computing and Data Management
- Respiratory viral infections research
- Advanced Image and Video Retrieval Techniques
- Healthcare Systems and Technology
- Chronic Disease Management Strategies
- Research Data Management Practices
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Biomedical Text Mining and Ontologies
- Human Pose and Action Recognition
- Music and Audio Processing
- Advanced Optical Imaging Technologies
- Thermography and Photoacoustic Techniques
- Computer Graphics and Visualization Techniques
- Multimodal Machine Learning Applications
- Advanced Vision and Imaging
- Pneumonia and Respiratory Infections
- Pediatric health and respiratory diseases
- Health Systems, Economic Evaluations, Quality of Life
- Artificial Intelligence in Healthcare
- Machine Learning and Data Classification
- Astronomical Observations and Instrumentation
- Explainable Artificial Intelligence (XAI)
University of Ottawa
2021
University of Washington Medical Center
2021
North China University of Technology
2018-2021
Xinjiang University
2021
Intermountain Healthcare
2014-2020
Xiamen University of Technology
2018
Guangzhou Academy of Special Equipment Inspection and Testing
2015
University of Southern California
2015
University of Utah
2012-2014
Heidelberg University
2012
Background As a major chronic disease, asthma causes many emergency department (ED) visits and hospitalizations each year. Predictive modeling is key technology to prospectively identify high-risk asthmatic patients enroll them in care management for preventive reduce future hospital encounters, including inpatient stays ED visits. However, existing models predicting encounters are inaccurate. Usually, they miss over half of the who will incur incorrectly classify others not. This makes it...
Background Asthma is a major chronic disease that poses heavy burden on health care. To facilitate the allocation of care management resources aimed at improving outcomes for high-risk patients with asthma, we recently built machine learning model to predict asthma hospital visits in subsequent year asthma. Our more accurate than previous models. However, like most models, it offers no explanation its prediction results. This creates barrier use management, where interpretability desired....
Background: Both chronic obstructive pulmonary disease (COPD) and asthma incur heavy health care burdens.To support tailored preventive for these 2 diseases, predictive modeling is widely used to give warnings identify patients management.However, 3 gaps exist in current methods owing rarely factoring temporal aspects showing trends early change: (1) existing models seldom use features often late warnings, making reactive.A risk found at a relatively stage of declining health, when the poor...
Image captioning is a comprehensive task in computer vision (CV) and natural language processing (NLP). It can complete conversion from image to text, that is, the algorithm automatically generates corresponding descriptive text according input image. In this paper, we present an end-to-end model takes deep convolutional neural network (CNN) as encoder recurrent (RNN) decoder. order get better extraction, propose highly modularized multi-branch CNN, which could increase accuracy while...
Background: In children below the age of 2 years, bronchiolitis is most common reason for hospitalization. Each year in United States, causes 287,000 emergency department visits, 32%-40% which result Due to a lack evidence and objective criteria managing bronchiolitis, clinicians often make disposition decisions on hospitalization or discharge home subjectively, leading large practice variation. Our recent study provided first operational definition appropriate hospital admission patients...
Cell lines are frequently used as highly standardized and reproducible in vitro models for biomedical analyses assays. distributed by cell banks that operate databases describing their products. However, the description of lines' properties not across different banks. Existing line-related ontologies mostly focus on names, but do cover aspects like origin or optimal growth conditions. The objective this work is to develop an ontology allows a more comprehensive metadata, which should data...
The1 spot welding technique is widely used in the industrial production line, but it suffers inconsistent quality. Therefore, evaluation of spot-welding product great importance for production. Many destructive and nondestructive methods have been evaluation, they are inefficient hard to be applied mass In recent year, machine vision method has differentiate acceptable failed products according their solder joint images. This opened new opportunities quality using digital image technique....
The clinical research landscape has changed dramatically in recent years terms of both volume and complexity. This poses new challenges for Institutional Review Boards’ (IRBs) review efficiency quality, especially at large academic medical centers. article discusses the technical facets IRB modernization. We analyzed information technology used by IRBs institutions across United States. found that centers have a high electronic adoption rate; however, capabilities systems vary greatly....
Bronchiolitis is the leading cause of hospitalization in children under 2 years age. Each year United States, bronchiolitis results 287,000 emergency department visits, 32%-40% which end hospitalization. Frequently, disposition decisions (to discharge or hospitalize) are made subjectively because lack evidence and objective criteria for management, to significant practice variation, wasted health care use, suboptimal outcomes. At present, no operational definition appropriate hospital...
Research data and biospecimen repositories are valuable resources for biomedical investigators. Sharing these has great potential benefits including efficient use of resources, avoiding duplicate experiments, gathering adequate sample sizes, promoting collaboration. However, concerns from producers about difficulties getting proper acknowledgement their contributions increasingly becoming obstacles large-scale sharing in reality. In this research project we analyzed the inadequacy current...
As the field of medicine grows more complicated and doctors become specialized in a particular field, number healthcare providers involved healing an individual patient increases. This is particularly true patients with multiple chronic conditions. Establishing effective communications among care becomes critical to facilitate coordination efficient resource use, which will ultimately result health outcome improvement. The first step for understand who have been patient's their relationships...
Sharing data in biomedical research networks has great potential benefits including efficient use of resources, avoiding duplicate experiments and promoting collaboration. However, concerns from producers about difficulties getting proper acknowledgement for their contributions are becoming obstacles network wide sharing reality. Effective convenient ways intellectual property management acknowledging to the required. This paper analyzed system requirements a German liver cancer proposed...
The development of deep learning and neural networks has brought broad prospects to computer vision natural language processing. image captioning task combines cutting-edge methods in two fields. By building an end-to-end encoder-decoder model, its description performance can be greatly improved. In this paper, the multi-branch convolutional network is used as encoder extract features, recurrent generate descriptive text that matches input image. We conducted experiments on Flickr8k,...
Diffusion models have revolutionized the field of talking head generation, yet still face challenges in expressiveness, controllability, and stability long-time generation. In this research, we propose an EmotiveTalk framework to address these issues. Firstly, realize better control over generation lip movement facial expression, a Vision-guided Audio Information Decoupling (V-AID) approach is designed generate audio-based decoupled representations aligned with movements expression....