Academic Journal
Machine learning score to predict in-hospital outcomes in patients hospitalized in cardiac intensive care unit
| Τίτλος: | Machine learning score to predict in-hospital outcomes in patients hospitalized in cardiac intensive care unit |
|---|---|
| Συγγραφείς: | Weizman, Orianne, Hamzi, Kenza, Henry, Patrick, Schurtz, Guillaume, Hauguel-Moreau, Marie, Trimaille, Antonin, Bedossa, Marc, Dib, Jean Claude, Attou, Sabir, Boukertouta, Tanissia, Boccara, Franck, Pommier, Thibaut, Lim, Pascal, Bochaton, Thomas, Millischer, Damien, Merat, Benoit, Picard, Fabien, Grinberg, Nissim, Sulman, David, Pasdeloup, Bastien, El Ouahidi, Yassine, Gonçalves, Treçy, Vicaut, Eric, Dillinger, Jean-Guillaume, Toupin, Solenn, Pezel, Théo, Aboyans, Victor, Albert, Emeric, Albert, Franck, Alvain, Sean, Amri, Nabil, Andrieu, Stéphane, Auvray, Simon, Azzakani, Sonia, Azencot, Ruben, Boccara, Albert, Bonnefoy-Cudraz, Eric, Bonnet, Guillaume, Bouali, Nabil, Bouchot, Océane, Bouleti, Claire, Brette, Jean Baptiste, Canu, Marjorie, Chaib, Aures |
| Συνεισφορές: | Fondation Coeur et Recherche |
| Πηγή: | European Heart Journal - Digital Health ; volume 6, issue 2, page 218-227 ; ISSN 2634-3916 |
| Στοιχεία εκδότη: | Oxford University Press (OUP) |
| Έτος έκδοσης: | 2024 |
| Περιγραφή: | Aims Although some scores based on traditional statistical methods are available for risk stratification in patients hospitalized in cardiac intensive care units (CICUs), the interest of machine learning (ML) methods for risk stratification in this field is not well established. We aimed to build an ML model to predict in-hospital major adverse events (MAE) in patients hospitalized in CICU. Methods and results In April 2021, a French national prospective multicentre study involving 39 centres included all consecutive patients admitted to CICU. The primary outcome was in-hospital MAE, including death, resuscitated cardiac arrest, or cardiogenic shock. Using 31 randomly assigned centres as an index cohort (divided into training and testing sets), several ML models were evaluated to predict in-hospital MAE. The eight remaining centres were used as an external validation cohort. Among 1499 consecutive patients included (aged 64 ± 15 years, 70% male), 67 had in-hospital MAE (4.3%). Out of 28 clinical, biological, ECG, and echocardiographic variables, seven were selected to predict MAE in the training set (n = 844). Boosted cost-sensitive C5.0 technique showed the best performance compared with other ML methods [receiver operating characteristic area under the curve (AUROC) = 0.90, precision–recall AUC = 0.57, F1 score = 0.5]. Our ML score showed a better performance than existing scores (AUROC: ML score = 0.90 vs. Thrombolysis In Myocardial Infarction (TIMI) score: 0.56, Global Registry of Acute Coronary Events (GRACE) score: 0.52, Acute Heart Failure (ACUTE-HF) score: 0.65; all P < 0.05). Machine learning score also showed excellent performance in the external cohort (AUROC = 0.88). Conclusion This new ML score is the first to demonstrate improved performance in predicting in-hospital outcomes over existing scores in patients admitted to the intensive care unit based on seven simple and rapid clinical and echocardiographic variables. Trial Registration ClinicalTrials.gov Identifier: NCT05063097. |
| Τύπος εγγράφου: | article in journal/newspaper |
| Γλώσσα: | English |
| DOI: | 10.1093/ehjdh/ztae098 |
| DOI: | 10.1093/ehjdh/ztae098/61246007/ztae098.pdf |
| Διαθεσιμότητα: | https://doi.org/10.1093/ehjdh/ztae098 https://academic.oup.com/ehjdh/advance-article-pdf/doi/10.1093/ehjdh/ztae098/61246007/ztae098.pdf https://academic.oup.com/ehjdh/article-pdf/6/2/218/61246007/ztae098.pdf |
| Rights: | https://creativecommons.org/licenses/by/4.0/ |
| Αριθμός Καταχώρησης: | edsbas.E8CD30CB |
| Βάση Δεδομένων: | BASE |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://doi.org/10.1093/ehjdh/ztae098# Name: EDS - BASE (ns324271) Category: fullText Text: View record from BASE |
|---|---|
| Header | DbId: edsbas DbLabel: BASE An: edsbas.E8CD30CB RelevancyScore: 945 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 944.983581542969 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Machine learning score to predict in-hospital outcomes in patients hospitalized in cardiac intensive care unit – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Weizman%2C+Orianne%22">Weizman, Orianne</searchLink><br /><searchLink fieldCode="AR" term="%22Hamzi%2C+Kenza%22">Hamzi, Kenza</searchLink><br /><searchLink fieldCode="AR" term="%22Henry%2C+Patrick%22">Henry, Patrick</searchLink><br /><searchLink fieldCode="AR" term="%22Schurtz%2C+Guillaume%22">Schurtz, Guillaume</searchLink><br /><searchLink fieldCode="AR" term="%22Hauguel-Moreau%2C+Marie%22">Hauguel-Moreau, Marie</searchLink><br /><searchLink fieldCode="AR" term="%22Trimaille%2C+Antonin%22">Trimaille, Antonin</searchLink><br /><searchLink fieldCode="AR" term="%22Bedossa%2C+Marc%22">Bedossa, Marc</searchLink><br /><searchLink fieldCode="AR" term="%22Dib%2C+Jean+Claude%22">Dib, Jean Claude</searchLink><br /><searchLink fieldCode="AR" term="%22Attou%2C+Sabir%22">Attou, Sabir</searchLink><br /><searchLink fieldCode="AR" term="%22Boukertouta%2C+Tanissia%22">Boukertouta, Tanissia</searchLink><br /><searchLink fieldCode="AR" term="%22Boccara%2C+Franck%22">Boccara, Franck</searchLink><br /><searchLink fieldCode="AR" term="%22Pommier%2C+Thibaut%22">Pommier, Thibaut</searchLink><br /><searchLink fieldCode="AR" term="%22Lim%2C+Pascal%22">Lim, Pascal</searchLink><br /><searchLink fieldCode="AR" term="%22Bochaton%2C+Thomas%22">Bochaton, Thomas</searchLink><br /><searchLink fieldCode="AR" term="%22Millischer%2C+Damien%22">Millischer, Damien</searchLink><br /><searchLink fieldCode="AR" term="%22Merat%2C+Benoit%22">Merat, Benoit</searchLink><br /><searchLink fieldCode="AR" term="%22Picard%2C+Fabien%22">Picard, Fabien</searchLink><br /><searchLink fieldCode="AR" term="%22Grinberg%2C+Nissim%22">Grinberg, Nissim</searchLink><br /><searchLink fieldCode="AR" term="%22Sulman%2C+David%22">Sulman, David</searchLink><br /><searchLink fieldCode="AR" term="%22Pasdeloup%2C+Bastien%22">Pasdeloup, Bastien</searchLink><br /><searchLink fieldCode="AR" term="%22El+Ouahidi%2C+Yassine%22">El Ouahidi, Yassine</searchLink><br /><searchLink fieldCode="AR" term="%22Gonçalves%2C+Treçy%22">Gonçalves, Treçy</searchLink><br /><searchLink fieldCode="AR" term="%22Vicaut%2C+Eric%22">Vicaut, Eric</searchLink><br /><searchLink fieldCode="AR" term="%22Dillinger%2C+Jean-Guillaume%22">Dillinger, Jean-Guillaume</searchLink><br /><searchLink fieldCode="AR" term="%22Toupin%2C+Solenn%22">Toupin, Solenn</searchLink><br /><searchLink fieldCode="AR" term="%22Pezel%2C+Théo%22">Pezel, Théo</searchLink><br /><searchLink fieldCode="AR" term="%22Aboyans%2C+Victor%22">Aboyans, Victor</searchLink><br /><searchLink fieldCode="AR" term="%22Albert%2C+Emeric%22">Albert, Emeric</searchLink><br /><searchLink fieldCode="AR" term="%22Albert%2C+Franck%22">Albert, Franck</searchLink><br /><searchLink fieldCode="AR" term="%22Alvain%2C+Sean%22">Alvain, Sean</searchLink><br /><searchLink fieldCode="AR" term="%22Amri%2C+Nabil%22">Amri, Nabil</searchLink><br /><searchLink fieldCode="AR" term="%22Andrieu%2C+Stéphane%22">Andrieu, Stéphane</searchLink><br /><searchLink fieldCode="AR" term="%22Auvray%2C+Simon%22">Auvray, Simon</searchLink><br /><searchLink fieldCode="AR" term="%22Azzakani%2C+Sonia%22">Azzakani, Sonia</searchLink><br /><searchLink fieldCode="AR" term="%22Azencot%2C+Ruben%22">Azencot, Ruben</searchLink><br /><searchLink fieldCode="AR" term="%22Boccara%2C+Albert%22">Boccara, Albert</searchLink><br /><searchLink fieldCode="AR" term="%22Bonnefoy-Cudraz%2C+Eric%22">Bonnefoy-Cudraz, Eric</searchLink><br /><searchLink fieldCode="AR" term="%22Bonnet%2C+Guillaume%22">Bonnet, Guillaume</searchLink><br /><searchLink fieldCode="AR" term="%22Bouali%2C+Nabil%22">Bouali, Nabil</searchLink><br /><searchLink fieldCode="AR" term="%22Bouchot%2C+Océane%22">Bouchot, Océane</searchLink><br /><searchLink fieldCode="AR" term="%22Bouleti%2C+Claire%22">Bouleti, Claire</searchLink><br /><searchLink fieldCode="AR" term="%22Brette%2C+Jean+Baptiste%22">Brette, Jean Baptiste</searchLink><br /><searchLink fieldCode="AR" term="%22Canu%2C+Marjorie%22">Canu, Marjorie</searchLink><br /><searchLink fieldCode="AR" term="%22Chaib%2C+Aures%22">Chaib, Aures</searchLink> – Name: Author Label: Contributors Group: Au Data: Fondation Coeur et Recherche – Name: TitleSource Label: Source Group: Src Data: European Heart Journal - Digital Health ; volume 6, issue 2, page 218-227 ; ISSN 2634-3916 – Name: Publisher Label: Publisher Information Group: PubInfo Data: Oxford University Press (OUP) – Name: DatePubCY Label: Publication Year Group: Date Data: 2024 – Name: Abstract Label: Description Group: Ab Data: Aims Although some scores based on traditional statistical methods are available for risk stratification in patients hospitalized in cardiac intensive care units (CICUs), the interest of machine learning (ML) methods for risk stratification in this field is not well established. We aimed to build an ML model to predict in-hospital major adverse events (MAE) in patients hospitalized in CICU. Methods and results In April 2021, a French national prospective multicentre study involving 39 centres included all consecutive patients admitted to CICU. The primary outcome was in-hospital MAE, including death, resuscitated cardiac arrest, or cardiogenic shock. Using 31 randomly assigned centres as an index cohort (divided into training and testing sets), several ML models were evaluated to predict in-hospital MAE. The eight remaining centres were used as an external validation cohort. Among 1499 consecutive patients included (aged 64 ± 15 years, 70% male), 67 had in-hospital MAE (4.3%). Out of 28 clinical, biological, ECG, and echocardiographic variables, seven were selected to predict MAE in the training set (n = 844). Boosted cost-sensitive C5.0 technique showed the best performance compared with other ML methods [receiver operating characteristic area under the curve (AUROC) = 0.90, precision–recall AUC = 0.57, F1 score = 0.5]. Our ML score showed a better performance than existing scores (AUROC: ML score = 0.90 vs. Thrombolysis In Myocardial Infarction (TIMI) score: 0.56, Global Registry of Acute Coronary Events (GRACE) score: 0.52, Acute Heart Failure (ACUTE-HF) score: 0.65; all P < 0.05). Machine learning score also showed excellent performance in the external cohort (AUROC = 0.88). Conclusion This new ML score is the first to demonstrate improved performance in predicting in-hospital outcomes over existing scores in patients admitted to the intensive care unit based on seven simple and rapid clinical and echocardiographic variables. Trial Registration ClinicalTrials.gov Identifier: NCT05063097. – Name: TypeDocument Label: Document Type Group: TypDoc Data: article in journal/newspaper – Name: Language Label: Language Group: Lang Data: English – Name: DOI Label: DOI Group: ID Data: 10.1093/ehjdh/ztae098 – Name: DOI Label: DOI Group: ID Data: 10.1093/ehjdh/ztae098/61246007/ztae098.pdf – Name: URL Label: Availability Group: URL Data: https://doi.org/10.1093/ehjdh/ztae098<br />https://academic.oup.com/ehjdh/advance-article-pdf/doi/10.1093/ehjdh/ztae098/61246007/ztae098.pdf<br />https://academic.oup.com/ehjdh/article-pdf/6/2/218/61246007/ztae098.pdf – Name: Copyright Label: Rights Group: Cpyrght Data: https://creativecommons.org/licenses/by/4.0/ – Name: AN Label: Accession Number Group: ID Data: edsbas.E8CD30CB |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.E8CD30CB |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1093/ehjdh/ztae098 Languages: – Text: English Titles: – TitleFull: Machine learning score to predict in-hospital outcomes in patients hospitalized in cardiac intensive care unit Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Weizman, Orianne – PersonEntity: Name: NameFull: Hamzi, Kenza – PersonEntity: Name: NameFull: Henry, Patrick – PersonEntity: Name: NameFull: Schurtz, Guillaume – PersonEntity: Name: NameFull: Hauguel-Moreau, Marie – PersonEntity: Name: NameFull: Trimaille, Antonin – PersonEntity: Name: NameFull: Bedossa, Marc – PersonEntity: Name: NameFull: Dib, Jean Claude – PersonEntity: Name: NameFull: Attou, Sabir – PersonEntity: Name: NameFull: Boukertouta, Tanissia – PersonEntity: Name: NameFull: Boccara, Franck – PersonEntity: Name: NameFull: Pommier, Thibaut – PersonEntity: Name: NameFull: Lim, Pascal – PersonEntity: Name: NameFull: Bochaton, Thomas – PersonEntity: Name: NameFull: Millischer, Damien – PersonEntity: Name: NameFull: Merat, Benoit – PersonEntity: Name: NameFull: Picard, Fabien – PersonEntity: Name: NameFull: Grinberg, Nissim – PersonEntity: Name: NameFull: Sulman, David – PersonEntity: Name: NameFull: Pasdeloup, Bastien – PersonEntity: Name: NameFull: El Ouahidi, Yassine – PersonEntity: Name: NameFull: Gonçalves, Treçy – PersonEntity: Name: NameFull: Vicaut, Eric – PersonEntity: Name: NameFull: Dillinger, Jean-Guillaume – PersonEntity: Name: NameFull: Toupin, Solenn – PersonEntity: Name: NameFull: Pezel, Théo – PersonEntity: Name: NameFull: Aboyans, Victor – PersonEntity: Name: NameFull: Albert, Emeric – PersonEntity: Name: NameFull: Albert, Franck – PersonEntity: Name: NameFull: Alvain, Sean – PersonEntity: Name: NameFull: Amri, Nabil – PersonEntity: Name: NameFull: Andrieu, Stéphane – PersonEntity: Name: NameFull: Auvray, Simon – PersonEntity: Name: NameFull: Azzakani, Sonia – PersonEntity: Name: NameFull: Azencot, Ruben – PersonEntity: Name: NameFull: Boccara, Albert – PersonEntity: Name: NameFull: Bonnefoy-Cudraz, Eric – PersonEntity: Name: NameFull: Bonnet, Guillaume – PersonEntity: Name: NameFull: Bouali, Nabil – PersonEntity: Name: NameFull: Bouchot, Océane – PersonEntity: Name: NameFull: Bouleti, Claire – PersonEntity: Name: NameFull: Brette, Jean Baptiste – PersonEntity: Name: NameFull: Canu, Marjorie – PersonEntity: Name: NameFull: Chaib, Aures – PersonEntity: Name: NameFull: Fondation Coeur et Recherche IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 01 Type: published Y: 2024 Identifiers: – Type: issn-locals Value: edsbas – Type: issn-locals Value: edsbas.oa Titles: – TitleFull: European Heart Journal - Digital Health ; volume 6, issue 2, page 218-227 ; ISSN 2634-3916 Type: main |
| ResultId | 1 |