Covid-19: Early identification and monitoring of the 2019 novel Coronavirus (Covid-19) disease community spread

点击翻译按钮获取中文摘要

The lack of health care information technology to leverage Electronic Health Records information (EHR) into useful insights locks the benefits of EHR systems. For example, COVID-19 is an unexpected outbreak that causes infections of the nose, throat and lungs and may cause death. In Canada, there are approximately 55,000 COVID-19. COVID-19 presents a sophisticated challenge and burden to the Canadian health care system because of its unexpected nature and high probability of potential life-loss. Currently, there is a knowledge gap in the development of early patient identification strategies of COVID-19 disease on a larger scale without in-person examination of patients at hospitals, due to the lack of technology to enable the remote identification and monitoring of this disease. The objectives of this study are: 1) The development of innovative predictive analytics algorithms to analyze patient EHR with nasopharynx swabs data to find undiagnosed patients at risk of spreading COVID-19. This objective will be achieved by correctly extracting and integrating COVID-19 semantic information from the multiple heterogeneous datasets that exist in the Canadian EHR systems, develop new predictive algorithms utilizing the extracted information, develop a new algorithm to estimate the probability of a patient to spread the disease, and develop a dynamic clustering algorithm to cluster patients into homogenous groups of different COVID-19 spreading probability (e.g., Low, High), 2)The development of an innovative platform to integrate predictive analytics algorithms with different medical databases into a shared decision-making mechanism. The shared decision-making mechanism can be used by physicians and policymakers to share information and work toward decisions about how to control the spread of the disease.