In:
PLOS Computational Biology, Public Library of Science (PLoS), Vol. 18, No. 1 ( 2022-1-18), p. e1009778-
Abstract:
The clinical outcome of SARS-CoV-2 infection varies widely between individuals. Machine learning models can support decision making in healthcare by assessing fatality risk in patients that do not yet show severe signs of COVID-19. Most predictive models rely on static demographic features and clinical values obtained upon hospitalization. However, time-dependent biomarkers associated with COVID-19 severity, such as antibody titers, can substantially contribute to the development of more accurate outcome models. Here we show that models trained on immune biomarkers, longitudinally monitored throughout hospitalization, predicted mortality and were more accurate than models based on demographic and clinical data upon hospital admission. Our best-performing predictive models were based on the temporal analysis of anti-SARS-CoV-2 Spike IgG titers, white blood cell (WBC), neutrophil and lymphocyte counts. These biomarkers, together with C-reactive protein and blood urea nitrogen levels, were found to correlate with severity of disease and mortality in a time-dependent manner. Shapley additive explanations of our model revealed the higher predictive value of day post-symptom onset (PSO) as hospitalization progresses and showed how immune biomarkers contribute to predict mortality. In sum, we demonstrate that the kinetics of immune biomarkers can inform clinical models to serve as a powerful monitoring tool for predicting fatality risk in hospitalized COVID-19 patients, underscoring the importance of contextualizing clinical parameters according to their time post-symptom onset.
Type of Medium:
Online Resource
ISSN:
1553-7358
DOI:
10.1371/journal.pcbi.1009778
DOI:
10.1371/journal.pcbi.1009778.g001
DOI:
10.1371/journal.pcbi.1009778.g002
DOI:
10.1371/journal.pcbi.1009778.g003
DOI:
10.1371/journal.pcbi.1009778.g004
DOI:
10.1371/journal.pcbi.1009778.g005
DOI:
10.1371/journal.pcbi.1009778.g006
DOI:
10.1371/journal.pcbi.1009778.t001
DOI:
10.1371/journal.pcbi.1009778.t002
DOI:
10.1371/journal.pcbi.1009778.s001
DOI:
10.1371/journal.pcbi.1009778.s002
DOI:
10.1371/journal.pcbi.1009778.s003
DOI:
10.1371/journal.pcbi.1009778.s004
DOI:
10.1371/journal.pcbi.1009778.s005
DOI:
10.1371/journal.pcbi.1009778.s006
DOI:
10.1371/journal.pcbi.1009778.s007
DOI:
10.1371/journal.pcbi.1009778.s008
DOI:
10.1371/journal.pcbi.1009778.s009
DOI:
10.1371/journal.pcbi.1009778.s010
DOI:
10.1371/journal.pcbi.1009778.s011
DOI:
10.1371/journal.pcbi.1009778.s012
DOI:
10.1371/journal.pcbi.1009778.s013
DOI:
10.1371/journal.pcbi.1009778.s014
DOI:
10.1371/journal.pcbi.1009778.s015
DOI:
10.1371/journal.pcbi.1009778.s016
DOI:
10.1371/journal.pcbi.1009778.s017
DOI:
10.1371/journal.pcbi.1009778.s018
DOI:
10.1371/journal.pcbi.1009778.s019
DOI:
10.1371/journal.pcbi.1009778.s020
DOI:
10.1371/journal.pcbi.1009778.r001
DOI:
10.1371/journal.pcbi.1009778.r002
DOI:
10.1371/journal.pcbi.1009778.r003
DOI:
10.1371/journal.pcbi.1009778.r004
Language:
English
Publisher:
Public Library of Science (PLoS)
Publication Date:
2022
detail.hit.zdb_id:
2193340-6
Bookmarklink