Machine learning identifies ICU outcome predictors in a multicenter COVID-19 cohort

  1. Magunia, Harry ORCID logo
  2. Lederer, Simone
  3. Verbuecheln, Raphael
  4. Gilot, Bryant Joseph
  5. Koeppen, Michael
  6. Haeberle, Helene A.
  7. Mirakaj, Valbona
  8. Hofmann, Pascal
  9. Marx, Gernot
  10. Bickenbach, Johannes
  11. Nohe, Boris
  12. Lay, Michael
  13. Spies, Claudia
  14. Edel, Andreas
  15. Schiefenhövel, Fridtjof
  16. Rahmel, Tim
  17. Putensen, Christian
  18. Sellmann, Timur
  19. Koch, Thea
  20. Brandenburger, Timo
  21. Kindgen-Milles, Detlef
  22. Brenner, Thorsten
  23. Berger, Marc
  24. Zacharowski, Kai
  25. Adam, Elisabeth
  26. Posch, Matthias
  27. Moerer, Onnen
  28. Scheer, Christian S.
  29. Sedding, Daniel
  30. Weigand, Markus A.
  31. Fichtner, Falk
  32. Nau, Carla
  33. Prätsch, Florian
  34. Wiesmann, Thomas
  35. Koch, Christian
  36. Schneider, Gerhard
  37. Lahmer, Tobias
  38. Straub, Andreas
  39. Meiser, Andreas
  40. Weiss, Manfred
  41. Jungwirth, Bettina
  42. Wappler, Frank
  43. Meybohm, Patrick
  44. Herrmann, Johannes
  45. Malek, Nisar
  46. Kohlbacher, Oliver
  47. Biergans, Stephanie
  48. Rosenberger, Peter

Download
s13054-021-03720-4.pdf 2,48MB
WeightNameValue
1000 Titel
  • Machine learning identifies ICU outcome predictors in a multicenter COVID-19 cohort
1000 Autor/in
  1. Magunia, Harry |
  2. Lederer, Simone |
  3. Verbuecheln, Raphael |
  4. Gilot, Bryant Joseph |
  5. Koeppen, Michael |
  6. Haeberle, Helene A. |
  7. Mirakaj, Valbona |
  8. Hofmann, Pascal |
  9. Marx, Gernot |
  10. Bickenbach, Johannes |
  11. Nohe, Boris |
  12. Lay, Michael |
  13. Spies, Claudia |
  14. Edel, Andreas |
  15. Schiefenhövel, Fridtjof |
  16. Rahmel, Tim |
  17. Putensen, Christian |
  18. Sellmann, Timur |
  19. Koch, Thea |
  20. Brandenburger, Timo |
  21. Kindgen-Milles, Detlef |
  22. Brenner, Thorsten |
  23. Berger, Marc |
  24. Zacharowski, Kai |
  25. Adam, Elisabeth |
  26. Posch, Matthias |
  27. Moerer, Onnen |
  28. Scheer, Christian S. |
  29. Sedding, Daniel |
  30. Weigand, Markus A. |
  31. Fichtner, Falk |
  32. Nau, Carla |
  33. Prätsch, Florian |
  34. Wiesmann, Thomas |
  35. Koch, Christian |
  36. Schneider, Gerhard |
  37. Lahmer, Tobias |
  38. Straub, Andreas |
  39. Meiser, Andreas |
  40. Weiss, Manfred |
  41. Jungwirth, Bettina |
  42. Wappler, Frank |
  43. Meybohm, Patrick |
  44. Herrmann, Johannes |
  45. Malek, Nisar |
  46. Kohlbacher, Oliver |
  47. Biergans, Stephanie |
  48. Rosenberger, Peter |
1000 Erscheinungsjahr 2021
1000 Publikationstyp
  1. Artikel |
1000 Online veröffentlicht
  • 2021-08-17
1000 Erschienen in
1000 Quellenangabe
  • 25(1):295
1000 Copyrightjahr
  • 2021
1000 Lizenz
1000 Verlagsversion
  • https://doi.org/10.1186/s13054-021-03720-4 |
  • https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8370055/ |
1000 Publikationsstatus
1000 Sprache der Publikation
1000 Abstract/Summary
  • Background!#!Intensive Care Resources are heavily utilized during the COVID-19 pandemic. However, risk stratification and prediction of SARS-CoV-2 patient clinical outcomes upon ICU admission remain inadequate. This study aimed to develop a machine learning model, based on retrospective & prospective clinical data, to stratify patient risk and predict ICU survival and outcomes.!##!Methods!#!A Germany-wide electronic registry was established to pseudonymously collect admission, therapeutic and discharge information of SARS-CoV-2 ICU patients retrospectively and prospectively. Machine learning approaches were evaluated for the accuracy and interpretability of predictions. The Explainable Boosting Machine approach was selected as the most suitable method. Individual, non-linear shape functions for predictive parameters and parameter interactions are reported.!##!Results!#!1039 patients were included in the Explainable Boosting Machine model, 596 patients retrospectively collected, and 443 patients prospectively collected. The model for prediction of general ICU outcome was shown to be more reliable to predict 'survival'. Age, inflammatory and thrombotic activity, and severity of ARDS at ICU admission were shown to be predictive of ICU survival. Patients' age, pulmonary dysfunction and transfer from an external institution were predictors for ECMO therapy. The interaction of patient age with D-dimer levels on admission and creatinine levels with SOFA score without GCS were predictors for renal replacement therapy.!##!Conclusions!#!Using Explainable Boosting Machine analysis, we confirmed and weighed previously reported and identified novel predictors for outcome in critically ill COVID-19 patients. Using this strategy, predictive modeling of COVID-19 ICU patient outcomes can be performed overcoming the limitations of linear regression models. Trial registration 'ClinicalTrials' (clinicaltrials.gov) under NCT04455451.
1000 Sacherschließung
gnd 1206347392 COVID-19
lokal Critical care
lokal Female [MeSH]
lokal Outcome
lokal ARDS
lokal Aged [MeSH]
lokal Adult [MeSH]
lokal Humans [MeSH]
lokal Middle Aged [MeSH]
lokal COVID-19
lokal COVID-19/therapy [MeSH]
lokal Cohort Studies [MeSH]
lokal Critical Illness/therapy [MeSH]
lokal Male [MeSH]
lokal Prognostic models
lokal Emergency Service, Hospital [MeSH]
lokal Intensive Care Units [MeSH]
lokal Research
lokal Germany [MeSH]
lokal Electronic Health Records/statistics
lokal Machine Learning [MeSH]
lokal Critical Illness/epidemiology [MeSH]
lokal Outcome Assessment, Health Care [MeSH]
lokal COVID-19/epidemiology [MeSH]
1000 Liste der Beteiligten
  1. https://orcid.org/0000-0001-9576-6399|https://frl.publisso.de/adhoc/uri/TGVkZXJlciwgU2ltb25l|https://frl.publisso.de/adhoc/uri/VmVyYnVlY2hlbG4sIFJhcGhhZWw=|https://frl.publisso.de/adhoc/uri/R2lsb3QsIEJyeWFudCBKb3NlcGg=|https://frl.publisso.de/adhoc/uri/S29lcHBlbiwgTWljaGFlbA==|https://frl.publisso.de/adhoc/uri/SGFlYmVybGUsIEhlbGVuZSBBLg==|https://frl.publisso.de/adhoc/uri/TWlyYWthaiwgVmFsYm9uYQ==|https://frl.publisso.de/adhoc/uri/SG9mbWFubiwgUGFzY2Fs|https://frl.publisso.de/adhoc/uri/TWFyeCwgR2Vybm90|https://frl.publisso.de/adhoc/uri/Qmlja2VuYmFjaCwgSm9oYW5uZXM=|https://frl.publisso.de/adhoc/uri/Tm9oZSwgQm9yaXM=|https://frl.publisso.de/adhoc/uri/TGF5LCBNaWNoYWVs|https://frl.publisso.de/adhoc/uri/U3BpZXMsIENsYXVkaWE=|https://frl.publisso.de/adhoc/uri/RWRlbCwgQW5kcmVhcw==|https://frl.publisso.de/adhoc/uri/U2NoaWVmZW5ow7Z2ZWwsIEZyaWR0am9m|https://frl.publisso.de/adhoc/uri/UmFobWVsLCBUaW0=|https://frl.publisso.de/adhoc/uri/UHV0ZW5zZW4sIENocmlzdGlhbg==|https://frl.publisso.de/adhoc/uri/U2VsbG1hbm4sIFRpbXVy|https://frl.publisso.de/adhoc/uri/S29jaCwgVGhlYQ==|https://frl.publisso.de/adhoc/uri/QnJhbmRlbmJ1cmdlciwgVGltbw==|https://frl.publisso.de/adhoc/uri/S2luZGdlbi1NaWxsZXMsIERldGxlZg==|https://frl.publisso.de/adhoc/uri/QnJlbm5lciwgVGhvcnN0ZW4=|https://frl.publisso.de/adhoc/uri/QmVyZ2VyLCBNYXJj|https://frl.publisso.de/adhoc/uri/WmFjaGFyb3dza2ksIEthaQ==|https://frl.publisso.de/adhoc/uri/QWRhbSwgRWxpc2FiZXRo|https://frl.publisso.de/adhoc/uri/UG9zY2gsIE1hdHRoaWFz|https://frl.publisso.de/adhoc/uri/TW9lcmVyLCBPbm5lbg==|https://frl.publisso.de/adhoc/uri/U2NoZWVyLCBDaHJpc3RpYW4gUy4=|https://frl.publisso.de/adhoc/uri/U2VkZGluZywgRGFuaWVs|https://frl.publisso.de/adhoc/uri/V2VpZ2FuZCwgTWFya3VzIEEu|https://frl.publisso.de/adhoc/uri/RmljaHRuZXIsIEZhbGs=|https://frl.publisso.de/adhoc/uri/TmF1LCBDYXJsYQ==|https://frl.publisso.de/adhoc/uri/UHLDpHRzY2gsIEZsb3JpYW4=|https://frl.publisso.de/adhoc/uri/V2llc21hbm4sIFRob21hcw==|https://frl.publisso.de/adhoc/uri/S29jaCwgQ2hyaXN0aWFu|https://frl.publisso.de/adhoc/uri/U2NobmVpZGVyLCBHZXJoYXJk|https://frl.publisso.de/adhoc/uri/TGFobWVyLCBUb2JpYXM=|https://frl.publisso.de/adhoc/uri/U3RyYXViLCBBbmRyZWFz|https://frl.publisso.de/adhoc/uri/TWVpc2VyLCBBbmRyZWFz|https://frl.publisso.de/adhoc/uri/V2Vpc3MsIE1hbmZyZWQ=|https://frl.publisso.de/adhoc/uri/SnVuZ3dpcnRoLCBCZXR0aW5h|https://frl.publisso.de/adhoc/uri/V2FwcGxlciwgRnJhbms=|https://frl.publisso.de/adhoc/uri/TWV5Ym9obSwgUGF0cmljaw==|https://frl.publisso.de/adhoc/uri/SGVycm1hbm4sIEpvaGFubmVz|https://frl.publisso.de/adhoc/uri/TWFsZWssIE5pc2Fy|https://frl.publisso.de/adhoc/uri/S29obGJhY2hlciwgT2xpdmVy|https://frl.publisso.de/adhoc/uri/QmllcmdhbnMsIFN0ZXBoYW5pZQ==|https://frl.publisso.de/adhoc/uri/Um9zZW5iZXJnZXIsIFBldGVy
1000 Hinweis
  • DeepGreen-ID: 0c84df261dc14288b6fa948c96bd8bb9 ; metadata provieded by: DeepGreen (https://www.oa-deepgreen.de/api/v1/), LIVIVO search scope life sciences (http://z3950.zbmed.de:6210/livivo), Crossref Unified Resource API (https://api.crossref.org/swagger-ui/index.html), to.science.api (https://frl.publisso.de/), ZDB JSON-API (beta) (https://zeitschriftendatenbank.de/api/), lobid - Dateninfrastruktur für Bibliotheken (https://lobid.org/resources/search)
1000 Label
1000 Dateien
1000 Objektart article
1000 Beschrieben durch
1000 @id frl:6463943.rdf
1000 Erstellt am 2023-11-15T22:57:58.154+0100
1000 Erstellt von 322
1000 beschreibt frl:6463943
1000 Zuletzt bearbeitet Thu Nov 30 22:40:19 CET 2023
1000 Objekt bearb. Thu Nov 30 22:40:19 CET 2023
1000 Vgl. frl:6463943
1000 Oai Id
  1. oai:frl.publisso.de:frl:6463943 |
1000 Sichtbarkeit Metadaten public
1000 Sichtbarkeit Daten public
1000 Gegenstand von

View source