A5087739684- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare and Education
- Sepsis Diagnosis and Treatment
- Genetic Associations and Epidemiology
- Coronary Interventions and Diagnostics
- Genetic and phenotypic traits in livestock
- Clinical Reasoning and Diagnostic Skills
- Genetic Mapping and Diversity in Plants and Animals
- Surgical site infection prevention
- Cardiac Imaging and Diagnostics
- Statistical Methods in Clinical Trials
- Fault Detection and Control Systems
- Outdoor and Experiential Education
- Green IT and Sustainability
- Advanced X-ray and CT Imaging
- Recreation, Leisure, Wilderness Management
- Image and Video Quality Assessment
- Network Traffic and Congestion Control
- Atrial Fibrillation Management and Outcomes
- Healthcare Technology and Patient Monitoring
- Diabetes Management and Research
- Health Systems, Economic Evaluations, Quality of Life
- Diabetes Management and Education
- Surgical Simulation and Training
- Statistical Methods and Inference
University of Toronto
2020-2024
Hamilton Health Sciences
2020-2024
Public Health Ontario
2020-2024
Beijing Jiaotong University
2016
Xi'an Jiaotong University
2003
Timely identification of patients at a high risk clinical deterioration is key to prioritizing care, allocating resources effectively, and preventing adverse outcomes. Vital signs-based, aggregate-weighted early warning systems are commonly used predict the outcomes related cardiorespiratory instability sepsis, which strong predictors poor mortality. Machine learning models, can incorporate trends capture relationships among parameters that models cannot, have recently been showing promising...
Identification of gene-by-environment interactions (GxE) is crucial to understand the interplay environmental effects on complex traits. However, current methods evaluating GxE biobank-scale datasets have limitations. We introduce MonsterLM, a multiple linear regression method that does not rely model specification and provides unbiased estimates variance explained by GxE. demonstrate robustness MonsterLM through comprehensive genome-wide simulations using real genetic data from 325,989...
Abstract While novel oral anticoagulants are increasingly used to reduce risk of stroke in patients with atrial fibrillation, vitamin K antagonists such as warfarin continue be extensively for prevention across the world. effective reducing strokes, complex pharmacodynamics make it difficult use clinically, many experiencing under- and/or over- anticoagulation. In this study we employed a implementation deep reinforcement learning provide clinical decision support optimize time therapeutic...
<sec> <title>BACKGROUND</title> Invasive coronary angiography (ICA) is the gold standard in diagnosis of artery disease (CAD). Being invasive, it carries rare but serious risks including myocardial infarction, stroke, major bleeding and death. A large proportion elective outpatients undergoing ICA have non-obstructive CAD, highlighting suboptimal use this test. Coronary computed tomographic (CCTA) a non-invasive option that provides similar information with less risk, recommended as...
Conventional clinical risk scores and diagnostic algorithms are proving to be suboptimal in the prediction of obstructive coronary artery disease, contributing low yield invasive angiography. Machine learning could help better predict which patients would benefit from angiography vs other noninvasive modalities.To reduce patient cost healthcare system by improving through optimized outpatient selection.Retrospective analysis 12 years referral data a provincial cardiac registry, including all...
Health coaching is an emerging intervention that has been shown to improve clinical and patient-relevant outcomes for type 2 diabetes. Advances in artificial intelligence may provide avenue developing a more personalized, adaptive, cost-effective approach diabetes health coaching.
Surgical site infections (SSIs) occur frequently and impact patients health care systems. Remote surveillance of surgical wounds is currently limited by the need for manual assessment clinicians. Machine learning (ML)-based methods have recently been used to address various aspects postoperative wound healing process may be improve scalability cost-effectiveness remote assessment.
In recent years, video streaming has contribute 40% traffic in LTE wireless networks. However, battery been a bottleneck for user to enjoy the high quality since progress of technique is far less than advancement technique. How enable users given limited an urgent problem. this paper, we propose transmission mechanism which optimize chunk size HTTP by taking into account watching behavior and characteristics Numerical results shows our can save energy apparently compared with existing schemes.
<sec> <title>BACKGROUND</title> Timely identification of patients at a high risk clinical deterioration is key to prioritizing care, allocating resources effectively, and preventing adverse outcomes. Vital signs–based, aggregate-weighted early warning systems are commonly used predict the outcomes related cardiorespiratory instability sepsis, which strong predictors poor mortality. Machine learning models, can incorporate trends capture relationships among parameters that models cannot, have...
Abstract Current methods to evaluate gene-by-environment (GxE) interactions on biobank-scale datasets are limited. MonsterLM enables multiple linear regression genome-wide datasets, does not rely parameters specification and provides unbiased estimates of variance explained by GxE interaction effects. We applied the UK Biobank for eight blood biomarkers (N=325,991), identifying significant with waist-to-hip ratio five biomarkers, ranging from 0.11 0.58. 48% 94% can be attributed variants...