Towards an Artificial Intelligence Framework for Data-Driven Prediction of Coronavirus Clinical Severity

Abstract

The virus SARS-CoV2, which causes coronavirus disease (COVID-19) has become a pandemic and has spread to every inhabited continent. Given the increasing caseload, there is an urgent need to augment clinical skills in order to identify from among the many mild cases the few that will progress to critical illness. We present a first step towards building an artificial intelligence (AI) framework, with predictive analytics (PA) capabilities applied to real patient data, to provide rapid clinical decision-making support. COVID-19 has presented a pressing need as a) clinicians are still developing clinical acumen to this novel disease and b) resource limitations in a surging pandemic require difficult resource allocation decisions. The objectives of this research are: (1) to algorithmically identify the combinations of clinical characteristics of COVID-19 that predict outcomes, and (2) to develop a tool with AI capabilities that will predict patients at risk for more severe illness on initial presentation. The predictive models learn from historical data to help predict who will develop acute respiratory distress syndrome (ARDS), a severe outcome in COVID-19. Our results, based on data from two hospitals in Wenzhou, Zhejiang, China, identified features on initial presentation with COVID-19 that were most predictive of later development of ARDS. A mildly elevated alanine aminotransferase (ALT) (a liver enzyme), the presence of myalgias (body aches), and an elevated hemoglobin (red blood cells), in this order, are the clinical features, on presentation, that are the most predictive. The predictive models that learned from historical data of patients from these two hospitals achieved 70% to 80% accuracy in predicting severe cases.

COVID-19 diagnosis prediction in emergency care patients: a machine learning approach

“…9 Another study of 53 patients from two hospitals in Wenzhou, China, analyzed the accuracy of five machine learning algorithms to predict Acute Respiratory Distress Syndrome (ARDS) in patients with COVID-19. 10 Two other studies applied machine learning algorithms to predict mortality in patients with COVID-19, using patient data from Kaggle and China. 11,12 We propose that machine learning algorithms can be used to allocate priorities in receiving the RT-PCR tests in the case of a shortage, and also to help with critical care decisions while the RT-PCR results are being processed (which have been frequently taking more than a week in most places of Brazil).…”

A predictive model to estimate survival of hospitalized COVID-19 patients from admission data

“…Attempting to augment medical decision-making, studies ranging from modulating single parameters to advanced predictive modeling have been applied to forecast decompensation, mortality, and survival among other clinical outcomes. [24][25][26] Early work with small patient cohorts of COVID-19 has led to models that identify some clinical characteristics that can be applied to predict severe cases (Yan et al, 2020, Jiang et al, 2020. 27 28 However, these studies are limited to small numbers of patients as well as the inclusion of qualitative and subjective variables, are prone to mislabeling, and are not always readily available.…”

Artificial intelligence vs COVID-19: limitations, constraints and pitfalls

“…Yan et al (2020) used Machine Learning to develop a prognostic prediction algorithm to predict the mortality risk of a person that has been infected, using data from (only) 29 patients at Tongji Hospital in Wuhan, China. And Jiang et al (2020) presents an AI that can predict with 80% accuracy which person affected with COVID-19 may go on to develop acute respiratory distress syndrome (ARDS). The sample that they used to train their AI system is, however, small (only 53 patients) and restricted to two Chinese hospitals.…”