Intelligent health system for the investigation of consenting COVID-19 patients and precision medicine
Personalized Medicine, our partnered journal, has recently published a perspective exploring the use of an AI intelligent health system to analyse COVID-19 data. Employing an intelligent health system within data analysis will enable us to identify factors influencing COVID-19 diagnosis, recovery, risk factors and mortality, which in turn will lead to the development of more successful prevention and treatment methods.
Abstract
Advancing frontiers of clinical research, we discuss the need for intelligent health systems to support a deeper investigation of COVID-19. We hypothesize that the convergence of the healthcare data and staggering developments in artificial intelligence have the potential to elevate the recovery process with diagnostic and predictive analysis to identify major causes of mortality, modifiable risk factors and actionable information that supports the early detection and prevention of COVID-19. However, current constraints include the recruitment of COVID-19 patients for research; translational integration of electronic health records and diversified public datasets; and the development of artificial intelligence systems for data-intensive computational modeling to assist clinical decision making. We propose a novel nexus of machine learning algorithms to examine COVID-19 data granularity from population studies to subgroups stratification and ensure best modeling strategies within the data continuum.
The quest to comprehend COVID-19 is the central focus of humankind today [1]. We have now realized that to effectively diagnose and treat patients with COVID-19 symptoms, we need an evolving understanding of the complex nature and course of the disease. [2,3]. Healthcare data analysis has the potential to transform the healthcare sector by predicting vulnerabilities and tailoring COVID-19 prevention and treatment [4]. In the past decades, various systems have been developed in academic and commercial sectors, but both are unable to identify problems by their effects and significantly help in clinical decision making with the healthcare data analytics and timely academic research collaboration [5–7]. Our central hypothesis states that integrative, intelligent and analytic access to the healthcare data of high volume, velocity, variety and veracity has the potential to revolutionize the field of medicine by providing the best strategies to diagnose and treat COVID-19 patients, especially those at risk of serious medical complications arising from a better understanding of the biology [8]. However, current limitations imply a gap in clinical and academic settings, difficulties in getting exigent approvals and timeliness of data availability, levels of granularity in clinical information and application of appropriate modeling strategies that allow learning in the data continuum. To efficiently establish a COVID-19 investigation, we need to deal with the major barriers to successfully implementing artificial intelligence (AI) in the healthcare sector, which include devising patient-centric protocols to improve the enrolment rates; translational integration of electronic health records (EHRs) and diversified publicly available datasets with optimization tools [9]; and the development of a Health Insurance Portability and Accountability Act-compliant AI system for data-intensive computational modeling to assist clinical decision making [10].
