CoChat
Sign Up
Cited by 104Machine Learning in HealthcareArtificial Intelligence in Healthcare and EducationCOVID-19 diagnosis using AI

The DOI matches its Crossref record; 64 of the 64 references it cites were found in a registry or an index.

Machine learning to assist clinical decision-making during the COVID-19 pandemic

Open access
Debnath, Shubham, Barnaby, Douglas P., Coppa, Kevin, Makhnevich, Alexander, Kim, Eun Ji, Chatterjee, Saurav, Tóth, Viktor, Levy, Todd J., Paradis, Marc d., Cohen, Stuart L., Hirsch, Jamie S., Zanos, Theodoros P., the Northwell COVID-19 Research Consortium, Becker, Lance B., Cookingham, Jennifer, Davidson, Karina W., Dominello, Andrew J., Falzon, Louise, McGinn, Thomas, Mogavero, Jazmin N., Osorio, Gabrielle A.·2020·Bioelectronic Medicine
DOI 10.1186/s42234-020-00050-8PMC PMC7347420
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: The number of cases from the coronavirus disease 2019 (COVID-19) global pandemic has overwhelmed existing medical facilities and forced clinicians, patients, and families to make pivotal decisions with limited time and information. Main body: While machine learning (ML) methods have been previously used to augment clinical decisions, there is now a demand for “Emergency ML.” Throughout the patient care pathway, there are opportunities for ML-supported decisions based on collected vitals, laboratory results, medication orders, and comorbidities. With rapidly growing datasets, there also remain important considerations when developing and validating ML models. Conclusion: This perspective highlights the utility of evidence-based prediction tools in a number of clinical settings, and how similar models can be deployed during the COVID-19 pandemic to guide hospital frontlines and healthcare administrators to make informed decisions about patient care and managing hospital volume.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s42234-020-00050-8

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 a ML/AI in the patient care pathway. The black asterisks represent multiple decision points during the patient care pathway that could be augmented by ML/AI tools. The green traces represent a COVID-19 negative diagnosis or recovery while the orange and red traces represent risk stratification of patients by lower and higher risks of deterioration, respectively, as determined by a potential ML/AI model. Additional decisions in the hospital include prioritization of care, allocation of resources, and estimation of prognosis. b An expanding COVID-19 database. Since March 1, 2020, there has been an increasing amount of COVID-19 patient data, shown here by new admissions and new medical data entries at Northwell Health, New York’s largest health system, facilities. Given increasing hospital admissions (black trace, left y-axis), there have been hundreds of thousands of new data entries per day (colored bars, right y-axis), including vitals, laboratory results, medication orders, and patient comorbidities. This vast data allows a unique opportunity to implement ML/AI to support medical frontlines and healthcare administrators in the fight against COVID-19. c Evolving patient profiles and discharge rates. Basic characteristics of the patient population changed during the progression of the wave of new cases, which can affect performance of a predictive model. For example, the average length of stay for expired patients and those discharged alive (dark blue with square markers and light blue with triangle markers, respectively, left y-axis) diverged in mid-April. Because of these changes, a predictive model with good early performance may decline because of differences between patients hospitalized for three weeks compared to less than a week. Also, an individual’s patient profile may have evolved significantly from hospital admission to those timepoints later during hospitalization. Discharges per day (grey bars, right y-axis) increased with the pandemic’s peak and declined with reduced numbers of new cases
Fig. 1 a ML/AI in the patient care pathway. The black asterisks represent multiple decision points during the patient care pathway that could be augmented by ML/AI tools. The green traces represent a COVID-19 negative diagnosis or recovery while the orange and red traces represent risk stratification of patients by lower and higher risks of deterioration, respectively, as determined by a potential ML/AI model. Additional decisions in the hospital include prioritization of care, allocation of resources, and estimation of prognosis. b An expanding COVID-19 database. Since March 1, 2020, there has been an increasing amount of COVID-19 patient data, shown here by new admissions and new medical data entries at Northwell Health, New York’s largest health system, facilities. Given increasing hospital admissions (black trace, left y-axis), there have been hundreds of thousands of new data entries per day (colored bars, right y-axis), including vitals, laboratory results, medication orders, and patient comorbidities. This vast data allows a unique opportunity to implement ML/AI to support medical frontlines and healthcare administrators in the fight against COVID-19. c Evolving patient profiles and discharge rates. Basic characteristics of the patient population changed during the progression of the wave of new cases, which can affect performance of a predictive model. For example, the average length of stay for expired patients and those discharged alive (dark blue with square markers and light blue with triangle markers, respectively, left y-axis) diverged in mid-April. Because of these changes, a predictive model with good early performance may decline because of differences between patients hospitalized for three weeks compared to less than a week. Also, an individual’s patient profile may have evolved significantly from hospital admission to those timepoints later during hospitalization. Discharges per day (grey bars, right y-axis) increased with the pandemic’s peak and declined with reduced numbers of new casesCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 57 references

Research works printed but not linked: 1

  1. Evans M, Armour S. Hospital capacity crosses tipping point in U.S. coronavirus hot spots. In: the wall street journal. New York: Dow Jones & Company, Inc.; 2020.

Cited by

Sorted by citations

Related

See all

Metadata

Authors
Debnath, S., Barnaby, D. P., Coppa, K., Makhnevich, A., Kim, E. J., Chatterjee, S., Tóth, V., Levy, T. J., Paradis, M. D., Cohen, S. L., Hirsch, J. S., Zanos, T. P., the Northwell COVID-19 Research Consortium, Becker, L. B., Cookingham, J., Davidson, K. W., Dominello, A. J., Falzon, L., McGinn, T., Mogavero, J. N., & Osorio, G. A.
Affiliations
Northwell Health (US)
Corresponding author
Zanos, Theodoros P.
ORCID
Venue
Bioelectronic Medicine
Publisher
Springer Science and Business Media LLC
Published
2020-07-10
Volume
6
Issue
1
Pages
14-14
ISSN
2332-8886 (ISSN-L)
DOI
10.1186/s42234-020-00050-8
PMC
PMC7347420
OpenAlex
W3042081405
Cited by
104
References
57
Citations by year
2026: 3 · 2025: 10 · 2024: 12 · 2023: 17 · 2022: 29 · 2021: 27 · 2020: 6
FWCI
7.76
Type
journal article
Language
en
Open access
Licence
Locations
  • Bioelectronic Medicine (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Institute on Aging (R24AG064191)
  • U.S. National Library of Medicine (R01LM012836)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
Keywords
COVID-19 pandemic, machine learning, clinical decision-making, clinical prediction model, model validation, hospital resource allocation, health care administrators, care pathways

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review