Microcirculatory alterations in critically ill COVID-19 patients analyzed using artificial intelligence

Critical Care, vol. 26, pp. 311

Abstract

BACKGROUND: The sublingual microcirculation presumably exhibits disease-specific changes in function and morphology. Algorithm-based quantification of functional microcirculatory hemodynamic variables in handheld vital microscopy (HVM) has recently allowed identification of hemodynamic alterations in the microcirculation associated with COVID-19. In the present study we hypothesized that supervised deep machine learning could be used to identify previously unknown microcirculatory alterations, and combination with algorithmically quantified functional variables increases the model's performance to differentiate critically ill COVID-19 patients from healthy volunteers. METHODS: Four international, multi-central cohorts of critically ill COVID-19 patients and healthy volunteers (n = 59/n = 40) were used for neuronal network training and internal validation, alongside quantification of functional microcirculatory hemodynamic variables. Independent verification of the models was performed in a second cohort (n = 25/n = 33). RESULTS: Six thousand ninety-two image sequences in 157 individuals were included. Bootstrapped internal validation yielded AUROC(CI) for detection of COVID-19 status of 0.75 (0.69-0.79), 0.74 (0.69-0.79) and 0.84 (0.80-0.89) for the algorithm-based, deep learning-based and combined models. Individual model performance in external validation was 0.73 (0.71-0.76) and 0.61 (0.58-0.63). Combined neuronal network and algorithm-based identification yielded the highest externally validated AUROC of 0.75 (0.73-0.78) (P < 0.0001 versus internal validation and individual models). CONCLUSIONS: We successfully trained a deep learning-based model to differentiate critically ill COVID-19 patients from heathy volunteers in sublingual HVM image sequences. Internally validated, deep learning was superior to the algorithmic approach. However, combining the deep learning method with an algorithm-based approach to quantify the functional state of the microcirculation markedly increased the sensitivity and specificity as compared to either approach alone, and enabled successful external validation of the identification of the presence of microcirculatory alterations associated with COVID-19 status.

Authors 13

1. Matthias P. Hilty corresponding

   University of Zurich · Erasmus MC · University Hospital Zurich

   Affiliation as printed

   Department of Intensive Care, Erasmus MC, University Medical Center, Rotterdam, The Netherlands. matthias.hilty@usz.ch

   Institute of Intensive Care Medicine, University Hospital of Zurich, Rämistrasse 100, 8091, Zurich, Switzerland. matthias.hilty@usz.ch

   Institute of Intensive Care Medicine, University Hospital of Zurich, Rämistrasse 100, 8091, Zurich, Switzerland

2. Emanuele Favaron

   Erasmus MC

   Affiliation as printed

   Department of Intensive Care, Erasmus MC, University Medical Center, Rotterdam, The Netherlands

3. Pedro David Wendel‐Garcia

   University of Zurich · University Hospital Zurich

   Affiliation as printed

   Institute of Intensive Care Medicine, University Hospital of Zurich, Rämistrasse 100, 8091, Zurich, Switzerland

4. Yavuz Ahiska
   Affiliation as printed

   Active Medical BV, Leiden, The Netherlands

5. Zühre Uz Aachen Department of Intensive Care

   Leiden University Medical Center

   Affiliation as printed

   Department of Intensive Care, Leiden University Medical Center, Leiden, The Netherlands

6. Şakir Akın

   Haga Hospital

   Affiliation as printed

   Department of Intensive Care, Haga Hospital, The Hague, The Netherlands

7. Moritz Flick

   University Medical Center Hamburg-Eppendorf

   Affiliation as printed

   Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

8. M. Sesmu Arbous Aachen Department of Intensive Care

   Leiden University Medical Center

   Affiliation as printed

   Department of Intensive Care, Leiden University Medical Center, Leiden, The Netherlands

9. Daniel A. Hofmaenner

   University of Zurich · University Hospital Zurich

   Affiliation as printed

   Institute of Intensive Care Medicine, University Hospital of Zurich, Rämistrasse 100, 8091, Zurich, Switzerland

10. Bernd Saugel

    University Medical Center Hamburg-Eppendorf

    Affiliation as printed

    Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

11. Henrik Endeman

    Erasmus MC

    Affiliation as printed

    Department of Intensive Care, Erasmus MC, University Medical Center, Rotterdam, The Netherlands

12. Reto Andreas Schuepbach

    University of Zurich · University Hospital Zurich

    Affiliation as printed

    Institute of Intensive Care Medicine, University Hospital of Zurich, Rämistrasse 100, 8091, Zurich, Switzerland

13. Can İnce

    Erasmus MC

    Affiliation as printed

    Department of Intensive Care, Erasmus MC, University Medical Center, Rotterdam, The Netherlands