ICU Patient Deterioration Prediction : A Data-Mining Approach

Abstract

A huge amount of medical data is generated every day, which presents a challenge in analysing these data. The obvious solution to this challenge is to reduce the amount of data without information loss. Dimension reduction is considered the most popular approach for reducing data size and also to reduce noise and redundancies in data. In this paper, we investigate the effect of feature selection in improving the prediction of patient deterioration in ICUs. We consider lab tests as features. Thus, choosing a subset of features would mean choosing the most important lab tests to perform. If the number of tests can be reduced by identifying the most important tests, then we could also identify the redundant tests. By omitting the redundant tests, observation time could be reduced and early treatment could be provided to avoid the risk. Additionally, unnecessary monetary cost would be avoided. Our approach uses state-ofthe-art feature selection for predicting ICU patient deterioration using the medical lab results. We apply our technique on the publicly available MIMIC-II database and show the effectiveness of the feature selection. We also provide a detailed analysis of the best features identified by our approach.

Prediction of ICU Patients’ Deterioration Using Machine Learning Techniques

“…To our knowledge, studies in the literature have yet to use these models in a similar dataset. Nuaimi et al [ 4 ] used the same dataset as ours. However, they solely employed medical laboratory testing and relied on feature selection to lower the dataset size in their technique.…”

“…The healthcare sector is transitioning from traditional practices to modern evidence-based care, primarily driven by data collected from sources such as electronic health records (EHRs) and monitoring devices [ 4 ]. Data mining explores large datasets to uncover valid, novel, and potentially useful patterns and relationships.…”

“…Previous studies have used data mining strategies to provide early deterioration alerts for ICU patients. For example, Mao et al developed an integrated data-mining approach using time-series features and various data-mining techniques [ 8 ], while Nuaimi et al focused on medical laboratory testing and feature selection to reduce dataset size [ 4 ]. In contrast, Liu et al created a search engine for health-related information and discovered that summarized cluster-based predictions outperformed non-summarized data [ 9 ].…”

A Novel and Reliable Framework of Patient Deterioration Prediction in Intensive Care Unit Based on Long Short-Term Memory-Recurrent Neural Network

“…The dataset is used to define training, validation, and testing data sets. Building solid models requires selecting the significant features that affect the performance of the model while implementing it [115]. Feature selection is performed to obviate overfitting and develop model performance.…”