VOOZH about

URL: https://pubmed.ncbi.nlm.nih.gov/33455840/

⇱ Predicting Progression to Septic Shock in the Emergency Department Using an Externally Generalizable Machine-Learning Algorithm - PubMed

Clipboard, Search History, and several other advanced features are temporarily unavailable.
Skip to main page content
👁 Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

👁 Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Abstract

Study objective: Machine-learning algorithms allow improved prediction of sepsis syndromes in the emergency department (ED), using data from electronic medical records. Transfer learning, a new subfield of machine learning, allows generalizability of an algorithm across clinical sites. We aim to validate the Artificial Intelligence Sepsis Expert for the prediction of delayed septic shock in a cohort of patients treated in the ED and demonstrate the feasibility of transfer learning to improve external validity at a second site.

Methods: This was an observational cohort study using data from greater than 180,000 patients from 2 academic medical centers between 2014 and 2019, using multiple definitions of sepsis. The Artificial Intelligence Sepsis Expert algorithm was trained with 40 input variables at the development site to predict delayed septic shock (occurring greater than 4 hours after ED triage) at various prediction windows. We then validated the algorithm at a second site, using transfer learning to demonstrate generalizability of the algorithm.

Results: We identified 9,354 patients with severe sepsis, of whom 723 developed septic shock at least 4 hours after triage. The Artificial Intelligence Sepsis Expert algorithm demonstrated excellent area under the receiver operating characteristic curve (>0.8) at 8 and 12 hours for the prediction of delayed septic shock. Transfer learning significantly improved the test characteristics of the Artificial Intelligence Sepsis Expert algorithm and yielded comparable performance at the validation site.

Conclusion: The Artificial Intelligence Sepsis Expert algorithm accurately predicted the development of delayed septic shock. The use of transfer learning allowed significantly improved external validity and generalizability at a second site. Future prospective studies are indicated to evaluate the clinical utility of this model.

PubMed Disclaimer

Figures

👁 Figure 1.
Figure 1.
Transfer learning as implemented in this study. In part 1, a prediction model neural network (such as Artificial Intelligence Sepsis Expert) is initialized with random weights and is trained with data from the development site. In part 2, starting with a pretrained model from the development site, the model weights are fine-tuned (or retrained) with a relatively small amount of data from the validation site before application to the validation cohort. This procedure is called transfer learning.
👁 Figure 2.
Figure 2.
Inclusion of subjects used in the development and validation cohorts using CMS sepsis definitions.
👁 Figure 3.
Figure 3.
Comparison of AUC and specificity (calculated at 85% sensitivity) of models predicting septic shock at different prediction windows at the validation site. Ten-fold cross validation was performed and median and interquartile ranges are presented with lines and shaded areas, respectively.
👁 Figure 4.
Figure 4.
AUC ROC and precision-recall curves for predicting t-sepsis 12 hours in advance at the development site using CMS severe sepsis definition. AUCroc, Area under the receiver operating characteristic; AUCpr, AUC precision-recall curve.
👁 Figure 5.
Figure 5.
AUC ROC of the ability of the Artificial Intelligence Sepsis Expert algorithm to detect septic shock 12 hours ahead of time in the validation cohort with and without transfer learning (red and black dashed lines, respectively) based on increasing amounts of patient encounters in model development.

Update of

References

    1. Rhee C, Dantes R, Epstein L, et al. Incidence and trends of sepsis in US hospitals using clinical vs claims data, 2009-2014. JAMA. 2017;318:1241–1249. - PMC - PubMed
    1. Villar J, Clement JP, Stotts J, et al. Many emergency department patients with severe sepsis and septic shock do not meet diagnostic criteria within 3 hours of arrival. Ann Emerg Med. 2014;64:48–54. - PubMed
    1. Capp R, Horton CL, Takhar SS, et al. Predictors of patients who present to the emergency department with sepsis and progress to septic shock between 4 and 48 hours of emergency department arrival. Crit Care Med. 2015;43:983–988. - PubMed
    1. Wardi G, Wali AR, Villar J, et al. Unexpected intensive care transfer of admitted patients with severe sepsis. J Intensive Care. 2017;5:43. - PMC - PubMed
    1. Sakr Y, Vincent JL, Schuerholz T, et al. Early- versus late-onset shock in European intensive care units. Shock. 2007;28:636–643. - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources

Full text links
👁 Elsevier Science full text link
Elsevier Science Free PMC article
Cite
Send To

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.