Data processing pipeline for cardiogenic shock prediction using machine learning.

Introduction: Recent 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. Methods: We 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. Results: We achieved good classification performance thanks to data cleaning and imputation (cross-validated mean area under the curve 0.805) without hyperparameter optimization. Conclusion: We believe our pre-processing pipeline would prove helpful also for other classification and regression experiments.

Development and validation of DNA methylation scores in two European cohorts augment 10-year risk prediction of type 2 diabetes.

Type 2 diabetes mellitus (T2D) presents a major health and economic burden that could be alleviated with improved early prediction and intervention. While standard risk factors have shown good predictive performance, we show that the use of blood-based DNA methylation information leads to a significant improvement in the prediction of 10-year T2D incidence risk. Previous studies have been largely constrained by linear assumptions, the use of cytosine-guanine pairs one-at-a-time and binary outcomes. We present a flexible approach (via an R package, MethylPipeR) based on a range of linear and tree-ensemble models that incorporate time-to-event data for prediction. Using the Generation Scotland cohort (training set ncases = 374, ncontrols = 9,461; test set ncases = 252, ncontrols = 4,526) our best-performing model (area under the receiver operating characteristic curve (AUC) = 0.872, area under the precision-recall curve (PRAUC) = 0.302) showed notable improvement in 10-year onset prediction beyond standard risk factors (AUC = 0.839, precision-recall AUC = 0.227). Replication was observed in the German-based KORA study (n = 1,451, ncases = 142, P = 1.6 × 10-5).

Reorganization of an Emergency Medical System in a Mixed Urban-Rural Area.

The reorganization of an emergency medical system means that we look for new locations of ambulance stations with the aim of improving the accessibility of the service. We applied two tools that are well known in the operations research community, namely mathematical programming, and computer simulation. Using the hierarchical pq-median model, we proposed optimal locations of the stations throughout the country and within large towns. Several solutions have been calculated that differ in the number of stations that are supposed to be relocated to new positions. The locations proposed by the mathematical programming model were evaluated via computer simulation. The approach was demonstrated under the conditions of the Slovak Republic using real historical data on ambulance dispatches. We have concluded that (i) the distribution of the stations proposed by the hierarchical pq-median model overcomes the current distribution; the performance of the system has significantly improved even if only 10% of the stations are relocated to new municipalities; (ii) the variant that relocates 40% of the stations is a reasonable compromise between the benefits and induced costs; (iii) optimizing station locations in big towns can significantly improve the local as well as the nationwide performance indicators; the response times in two regional capitals has reduced by more than 4 min.