A robust and parsimonious machine learning method to predict ICU admission of COVID-19 patients

Abstract

In this article, we discuss the development of prognostic machine learning (ML) models for COVID-19 progression, by focusing on the task of predicting ICU admission within (any of) the next 5 days. On the basis of 6,625 complete blood count (CBC) tests from 1,004 patients, of which 18% were admitted to intensive care unit (ICU), we created four ML models, by adopting a robust development procedure which was designed to minimize risks of bias and over-fitting, according to reference guidelines. The best model, a support vector machine, had an AUC of .85, a Brier score of .14, and a standardized net benefit of .69: these scores indicate that the model performed well over a variety of prediction criteria. We also conducted an interpretability study to back up our findings, showing that the data on which the developed model is based is consistent with the current medical literature. This also demonstrates that CBC data and ML methods can be used to predict COVID-19 patients’ ICU admission at a relatively low cost: in particular, since CBC data can be quickly obtained by means of routine blood exams, our models could be used in resource-constrained settings and provide health practitioners with rapid and reliable indications. Graphical abstract

Machine learning models for predicting severe COVID-19 outcomes in hospitals

“…Famiglini et al [ 32 ] devised several machine-learning models predicting ICU admission of COVID-19 patients within the next five days using gender, age and the complete blood count as potential features. Their best model achieves a ROC-AUC of 0.85 and a Brier score of 0.144.…”

Prognosing the risk of COVID-19 death through a machine learning-based routine blood panel: A retrospective study in Brazil

“…Currently, there are several studies around the world using machine learning tecnique based on different characteristics easily obtained from patients with COVID-19 [13] , [18] , [40] and different outcomes such as need for intensive care, mechanical ventilation and death, among others. Among these characteristics, laboratory test data have also been used in machine learning prediction models, rather than subjective data that could vary between geographic regions, ethnic characteristics, observers and institutions.…”

“…Fernandes et al (2021) developed a ML model to predict the risk of ICU admission, use of mechanical ventilation and/or death analysing a database of 1,040 patients from a hospital in Brazil, using the routine biomarkers lymphocytes, CRP and ferritin, together with Intensive Care Unit (ICU) scores [17] . Famiglini et al (2022) developed ML models for the prediction of ICU patient admission using only Complete Blood Count (CBC) data from Italy [18] , which has been further externally tested in databases from multiple countries [19] , but had the worst performance rates in the Brazilian ones. All of these studies aimed to predict the risk of ICU admission, and some of them used parameters other than routine blood biomarkers.…”

Hematopathology of Severe Acute Respiratory Syndrome Coronavirus 2 Infection and Coronavirus Disease-19

“…In contrast, Gabr et al found significantly more atypical monocytes in ICU patients 26 , while Pezeshki et al did not identify any statistically significant associations between blood findings and clinical course 21 . Importantly, several non-hematologic factors (e.g., age, gender), and other laboratory parameters (including ferritin, C-reactive protein, interleukin-6) are often associated with death, and machine learning-based algorithms have been employed for diagnosis and prediction of care needs and outcome 35 , 36 , 37 .…”