How Could Machine Learning Help Healthcare Informatics Predict Coronavirus?

How Could Machine Learning Help Healthcare Informatics Predict Coronavirus?

Kenneth David Strang
Copyright: © 2023 |Pages: 22
DOI: 10.4018/978-1-6684-5499-2.ch002
(Individual Chapters)
No Current Special Offers


This chapter differs from most healthcare informatics studies because the focus is on conceptual COVID-19 SARS-CoV2 (coronavirus) prediction rather than detection. The research question was how state-of-the-art informatics software could be used to detect coronavirus based on the analysis of hospital patient medical records. Healthcare practitioners need artificial intelligence (AI) software to predict what has not yet happened to prepare in advance. Therefore, this chapter proposes and tests a generic AI approach to predict first-time coronavirus infection for discharged hospital patients based on data collected from their medical records. This idea could allow healthcare informatics practitioners to leverage AI software to predict which patients will be more likely to become infected by specific viruses or diseases. The concept rather than the actual model is the most valuable outcome of the study.
Chapter Preview


This chapter discusses an applied example of how artificial intelligence (AI) software was used to predict a virus using coronavirus as a prototype, based on retrospective analysis of a large medical database. This introductory section will highlight key issues in the literature in order to establish the rationale underling the current study. Next the relevant literature is reviewed and then a practical example is explained by analyzing medical data to predict coronavirus in non-intensive care hospital patients. The results are discussed in terms of how AI could contribute to disease analysis and prediction – in terms of health informatics stakeholders. Thus, this chapter is intended for health informatics professionals, but other healthcare stakeholders would likely benefit from these findings including medical physicians, clinical researchers, healthcare methods researchers including students, and pharmaceutical companies.

The coronavirus is an excellent proof of concept case study because the pandemic impacted everyone in the world. Think back to the start of the pandemic and imagine you are a healthcare informatics staff assisting medical teams when numerous patients arrive to an already over-burdened hospital. Remember this mysterious coronavirus caused patients to die quickly as compared to common ICU patient illnesses. Furthermore, it spread rapidly and invisibly. Early on, in the coronavirus pandemic, there were large volumes of healthcare informatics data being shared but practitioners struggled with analyzing it for causes or predictions of patient morbidity and mortality. In fact, most COVID-19 analytic software breakthroughs were detection not prediction, accomplished by identifying particular congestion patterns from lung x-ray or CT images of sick patients (Kuchana et al., 2021; Li et al., 2020; Wang et al., 2021; Wen et al., 2020). At best this was affirmation of what was suspected namely COVID-19 infection, or in certain cases the lack of particular patterns instead confirmed other repertory viruses such as pneumonia.

AI has already been applied to predict viruses. Google Flu Trends was an interesting well-known case of AI being used to predict severe flu outbreaks around the world (Strang, 2021). However, there were good as well as bad results. Google uses AI programming to search for flu symptoms, remedies, and other related keywords from browser Internet searches to provide near-real-time estimates of flu activity in the United States and 24 other countries (Lazer, Kennedy, King, & Vespignani, 2014). The AI principle was to assume searches meant the user or someone in their residence was experiencing symptoms.

According to Lazer, Kennedy, King and Vespignani (2014), Google Flu Trends provided a remarkably accurate indicator of the flu cases in the United States between 2009 and 2011, which was significantly more accurate than the CDC predictions. However, Google Flu Trends was inaccurate thereafter for 2012–2013, more than twice as high as the CDC predictions of which the latter were accurate (Lazer et al., 2014). Thus, self-perceived user symptoms of the virus became a proxy for a medical doctor suspecting the virus after examining a patient. The problem was many users were paranoid, falsely assuming they had a virus and not trusting a single result so multiple searches were done by them and their friends. Mild user hysteria propagated searches for the flu, and inflated actual cases by many times.

Thus, although Google Flu Trends AI was an accurate forecast tool over several years, it became inaccurate because the predictions for vaccinations and doctor visits increased much more than actual need which subsequently cost health providers and medical facilities though resource waste. Therefore, healthcare informatics practitioners need more reliable methods to predict patient infections as well as virus outbreaks.

Key Terms in this Chapter

COVID-19 SARS CoV2: The common coronavirus causes in 2019 and forming a global pandemic lasting until approximately 2022 with minor outbreaks thereafter.

Complete Chapter List

Search this Book: