Data processing pipeline for cardiogenic shock prediction using machine learning

• DOI: 10.3389/fcvm.2023.1132680
• Publication Date: 2023-03-23T10:01:42Z
• Authors: Nikola Jajcay; Branislav Bezak; Amitai Segev; Shlomi Matetzky; Jana Jankova; Michael Spartalis; Mohammad El Tahlawi; Federico Guerra; Julian Friebel; Tharusan Thevathasan; Imrich Berta; Leo Pölzl; Felix Nägele; Edita Pogran; F. Aaysha Cader; Milana Jarakovic; Can Gollmann-Tepe...; Marta Kollarova; Katarina Petrikova; Otilia Tica; Konstantin A. Kry...; Guido Tavazzi; Carsten Skurk; Kurt Huber; Allan Böhm
• Keywords: Artificial intelligence; 330; Cardiology; 610; Health Professions; Cardiovascular Medicine; missing data imputation; Health Information Management; Artificial Intelligence; Health Sciences; Machine learning; Diseases of the circulatory (Cardiovascular) system; Clinical Event Prediction; Disease Risk Prediction; Cardiogenic shock; Internal medicine; Machine Learning in Healthcare and Medicine; cardiogenic shock; Deep Learning Applications in Healthcare; processing pipeline; Analysis of Electrocardiogram Signals; Predictive Modeling; Computer science; 3. Good health; prediction model; Myocardial infarction; Operating system; machine learning; classification; RC666-701; Computer Science; Physical Sciences; Signal Processing; Medicine; Heart Disease Prediction; Pipeline (software); Cardiology and Cardiovascular Medicine

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.