Knowledge Extraction From ICU Data Using Data Visualization

Abstract

Due to ICU critical environment, where seriously ill patients must be constantly monitored, it is imperative to make quick but assertive decisions. Several studies have shown that continuous monitorization of ICU patients result in large amounts of data, from which knowledge can be extracted and better decisions made. This chapter aims to analyse and visualize the data obtained by an ICU, so that conclusions can be deduced regarding patients' outcome, clinical errors, as well as healthcare service quality. To achieve the objective, initially, the data was acquired and collected from several data sources such as bedside monitors and electronic nursing records. Secondly, the raw data was transformed so that it could be used in visualization. Finally, interactive charts were built so that data could be forecasted and patterns discovered. The results allow one to draw conclusions such as the source of data gaps, the correlation between medication and vital signs, as well as the importance of SAPS regarding patient outcomes.