Frontiers in Cardiovascular Medicine (Mar 2023)
Data processing pipeline for cardiogenic shock prediction using machine learning
- Nikola Jajcay,
- Nikola Jajcay,
- Branislav Bezak,
- Branislav Bezak,
- Branislav Bezak,
- Amitai Segev,
- Amitai Segev,
- Shlomi Matetzky,
- Shlomi Matetzky,
- Jana Jankova,
- Michael Spartalis,
- Michael Spartalis,
- Mohammad El Tahlawi,
- Federico Guerra,
- Julian Friebel,
- Tharusan Thevathasan,
- Tharusan Thevathasan,
- Tharusan Thevathasan,
- Tharusan Thevathasan,
- Imrich Berta,
- Leo Pölzl,
- Felix Nägele,
- Edita Pogran,
- F. Aaysha Cader,
- Milana Jarakovic,
- Milana Jarakovic,
- Can Gollmann-Tepeköylü,
- Marta Kollarova,
- Katarina Petrikova,
- Otilia Tica,
- Otilia Tica,
- Konstantin A. Krychtiuk,
- Konstantin A. Krychtiuk,
- Guido Tavazzi,
- Guido Tavazzi,
- Carsten Skurk,
- Carsten Skurk,
- Kurt Huber,
- Allan Böhm,
- Allan Böhm,
- Allan Böhm
Affiliations
- Nikola Jajcay
- Premedix Academy, Bratislava, Slovakia
- Nikola Jajcay
- Department of Complex Systems, Institute of Computer Science, Czech Academy of Sciences, Prague, Czech Republic
- Branislav Bezak
- Premedix Academy, Bratislava, Slovakia
- Branislav Bezak
- Clinic of Cardiac Surgery, National Institute of Cardiovascular Diseases, Bratislava, Slovakia
- Branislav Bezak
- Faculty of Medicine, Comenius University in Bratislava, Bratislava, Slovakia
- Amitai Segev
- The Leviev Cardiothoracic & Vascular Center, Chaim Sheba Medical Center, Ramat Gan, Israel
- Amitai Segev
- Affiliated to the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
- Shlomi Matetzky
- The Leviev Cardiothoracic & Vascular Center, Chaim Sheba Medical Center, Ramat Gan, Israel
- Shlomi Matetzky
- Affiliated to the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
- Jana Jankova
- Premedix Academy, Bratislava, Slovakia
- Michael Spartalis
- 3rd Department of Cardiology, National and Kapodistrian University of Athens, Athens, Greece
- Michael Spartalis
- Global Clinical Scholars Research Training (GCSRT) Program, Harvard Medical School, Boston, MA, United States
- Mohammad El Tahlawi
- Department of Cardiology, Faculty of Human Medicine, Zagazig University, Zagazig, Egypt
- Federico Guerra
- 0Cardiology and Arrhythmology Clinic, Marche Polytechnic University, University Hospital “Umberto I - Lancisi - Salesi”, Ancona, Italy
- Julian Friebel
- 1Department of Cardiology Angiology and Intensive Care Medicine, Deutsches Herzzentrum der Charité (DHZC), Campus Benjamin Franklin, Charité - Universitätsmedizin Berlin, Berlin, Germany
- Tharusan Thevathasan
- 1Department of Cardiology Angiology and Intensive Care Medicine, Deutsches Herzzentrum der Charité (DHZC), Campus Benjamin Franklin, Charité - Universitätsmedizin Berlin, Berlin, Germany
- Tharusan Thevathasan
- 2Berlin Institute of Health, Charité—Universitätsmedizin Berlin, Berlin, Germany
- Tharusan Thevathasan
- 3Deutsches Zentrum für Herz-Kreislauf-Forschung e.V., Berlin, Germany
- Tharusan Thevathasan
- 4Institute of Medical Informatics, Charité—Universitätsmedizin Berlin, Berlin, Germany
- Imrich Berta
- Premedix Academy, Bratislava, Slovakia
- Leo Pölzl
- 5Department for Cardiac Surgery, Cardiac Regeneration Research, Medical University of Innsbruck, Innsbruck, Austria
- Felix Nägele
- 5Department for Cardiac Surgery, Cardiac Regeneration Research, Medical University of Innsbruck, Innsbruck, Austria
- Edita Pogran
- 63rd Medical Department, Cardiology and Intensive Care Medicine, Wilhelminen Hospital, Vienna, Austria
- F. Aaysha Cader
- 7Department of Cardiology, Ibrahim Cardiac Hospital & Research Institute, Dhaka, Bangladesh
- Milana Jarakovic
- 8Cardiac Intensive Care Unit, Institute for Cardiovascular Diseases of Vojvodina, Sremska Kamenica, Serbia
- Milana Jarakovic
- 9Faculty of Medicine, University of Novi Sad, Novi Sad, Serbia
- Can Gollmann-Tepeköylü
- 5Department for Cardiac Surgery, Cardiac Regeneration Research, Medical University of Innsbruck, Innsbruck, Austria
- Marta Kollarova
- Premedix Academy, Bratislava, Slovakia
- Katarina Petrikova
- Premedix Academy, Bratislava, Slovakia
- Otilia Tica
- 0Cardiology Department, Emergency County Clinical Hospital of Oradea, Oradea, Romania
- Otilia Tica
- 1Institute of Cardiovascular Sciences, University of Birmingham, Medical School, Birmingham, United Kingdom
- Konstantin A. Krychtiuk
- 2Department of Internal Medicine II, Division of Cardiology, Medical University of Vienna, Vienna, Austria
- Konstantin A. Krychtiuk
- 3Duke Clinical Research Institute Durham, NC, United States
- Guido Tavazzi
- 4Department of Clinical-Surgical, Diagnostic and Paediatric Sciences, University of Pavia, Pavia, Italy
- Guido Tavazzi
- 5Anesthesia and Intensive Care, Fondazione Policlinico San Matteo Hospital IRCCS, Pavia, Italy
- Carsten Skurk
- 1Department of Cardiology Angiology and Intensive Care Medicine, Deutsches Herzzentrum der Charité (DHZC), Campus Benjamin Franklin, Charité - Universitätsmedizin Berlin, Berlin, Germany
- Carsten Skurk
- 3Deutsches Zentrum für Herz-Kreislauf-Forschung e.V., Berlin, Germany
- Kurt Huber
- 63rd Medical Department, Cardiology and Intensive Care Medicine, Wilhelminen Hospital, Vienna, Austria
- Allan Böhm
- Premedix Academy, Bratislava, Slovakia
- Allan Böhm
- Faculty of Medicine, Comenius University in Bratislava, Bratislava, Slovakia
- Allan Böhm
- 6Department of Acute Cardiology, National Institute of Cardiovascular Diseases, Bratislava, Slovakia
- DOI
- https://doi.org/10.3389/fcvm.2023.1132680
- Journal volume & issue
-
Vol. 10
Abstract
IntroductionRecent advances in machine learning provide new possibilities to process and analyse observational patient data to predict patient outcomes. In this paper, we introduce a data processing pipeline for cardiogenic shock (CS) prediction from the MIMIC III database of intensive cardiac care unit patients with acute coronary syndrome. The ability to identify high-risk patients could possibly allow taking pre-emptive measures and thus prevent the development of CS.MethodsWe mainly focus on techniques for the imputation of missing data by generating a pipeline for imputation and comparing the performance of various multivariate imputation algorithms, including k-nearest neighbours, two singular value decomposition (SVD)—based methods, and Multiple Imputation by Chained Equations. After imputation, we select the final subjects and variables from the imputed dataset and showcase the performance of the gradient-boosted framework that uses a tree-based classifier for cardiogenic shock prediction.ResultsWe achieved good classification performance thanks to data cleaning and imputation (cross-validated mean area under the curve 0.805) without hyperparameter optimization.ConclusionWe believe our pre-processing pipeline would prove helpful also for other classification and regression experiments.
Keywords
- classification
- machine learning
- missing data imputation
- processing pipeline
- prediction model
- cardiogenic shock