Machine Learning-Based Pressure Ulcer Prediction in Modular Critical Care Data

Abstract

Increasingly available open medical and health datasets encourage data-driven research with a promise of improving patient care through knowledge discovery and algorithm development. Among efficient approaches to such high-dimensional problems are a number of machine learning methods, which are applied in this paper to pressure ulcer prediction in modular critical care data. An inherent property of many health-related datasets is a high number of irregularly sampled time-variant and scarcely populated features, often exceeding the number of observations. Although machine learning methods are known to work well under such circumstances, many choices regarding model and data processing exist. In particular, this paper address both theoretical and practical aspects related to the application of six classification models to pressure ulcers, while utilizing one of the largest available Medical Information Mart for Intensive Care (MIMIC-IV) databases. Random forest, with an accuracy of 96%, is the best-performing approach among the considered machine learning algorithms.

An Integrated System of Braden Scale and Random Forest Using Real-Time Diagnoses to Predict When Hospital-Acquired Pressure Injuries (Bedsores) Occur

“…Sixty variables that include the Braden Risk Assessment subscales were used as inputs for a model to predict HAPI timing. The HAPI timing was identified as the number of days to develop HAPIs [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 46 ]. These variables are shown in Table 1 and selected through literature survey and clinician’s feedback [ 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 <...…”

“…Machine Learning (ML) has been utilized to predict HAPI patients before occurrence using patients’ Electronic Health Records (EHR) as an adjunct to clinical assessment, which can further narrow down which patients are at risk [ 6 , 7 , 8 , 9 ]. In the last 15 years, 30 studies have answered who will develop HAPIs by utilizing classic machine and deep learning algorithms [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 ]. Two studies adopted Grid Search (GS) to optimize the hyperparameters of ML [ 10 , 22 ].…”

“…Two studies adopted Grid Search (GS) to optimize the hyperparameters of ML [ 10 , 22 ]. In most studies, random oversampling, undersampling, and Synthetic Minority Oversampling Techniques (SMOTE) were adopted as oversampling techniques [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 38 ]. Nevertheless, one study adopted cost-sensitive learning [ 22 ] to deal with the unbalanced dataset.…”

“…Nevertheless, one study adopted cost-sensitive learning [ 22 ] to deal with the unbalanced dataset. The most used algorithms were Random Forest (RF) [ 10 , 12 , 13 , 14 , 15 , 16 , 19 , 21 , 22 , 25 , 30 , 31 , 32 , 36 , 38 , 39 ], Logistic Regression (LR) [ 10 , 11 , 14 , 16 , 18 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 29 , 30 , 31 , 32 , 34 , 36 ], Decision Tree (DT) [ 15 , 16 , 21 , 23 , 25 , 26 , 27 , 28 , 31 , 32 , 36 ], Support Vector Machine (SVM) [ 10 , 12 , 21 , 22 , 25 , 26 , 27 , 28 , 30 , ...…”

Deciphering the endometrial immune landscape of RIF during the window of implantation from cellular senescence by integrated bioinformatics analysis and machine learning

“…A tremendous amount of high-throughput data is piling up in public databases, which greatly helps with uncovering the potential etiopathogenesis and identifying candidate targets for drug design ( 22 ). Machine learning has recently been widely applied to learn the representation of high-dimensional features derived from gene expression data on account of its powerful capabilities in classification ( 23 , 24 ). The ingenious combination of bioinformatics analysis and machine learning is a creative and crucial way to establish novel diagnostic models and understand pathological mechanisms at the molecular level, which is in line with the latest research trends.…”