Advances in statistical methodology and their application in critical care

Abstract

By becoming familiar with advances in statistical methodology, researchers and clinicians can enhance collaboration with their statistical colleagues, toward the goal of better study design and analysis.

Mining data from intensive care patients

“…The growing number of papers in the critical care literature that surveys basic data mining methods ( (Lucas, 2004), (Kreke et al, 2004), (Kong et al, 2004), (Sierra et al, 2001)) shows the significant interest of researchers in this domain to apply data mining methods to improve their knowledge.…”

“…There is already a lot of literature on data mining for ICU. On the one hand there are articles aimed at medical doctors and other experts in intensive care, explaining the range of data mining methods that are currently available and/or illustrating on specific problems how they can be useful (Lucas, 2004;Kreke et al, 2004;Kong et al, 2004;Sierra et al, 2001). On the other hand, there are articles aiming also (or mainly) at an audience from machine learning or statistics (Moser et al, 1999;Morik et al, 2000;Ganzert et al, 2002), introducing specific problems in intensive care and how they can be solved with data mining techniques.…”

Analysis of heart rate variability as a predictor of mortality in cardiovascular patients of intensive care unit

“…Some research used machine learning algorithms, such as artificial neural networks and decision trees as a prediction algorithm in different critical care settings [22][23][24][25][26][27][28]. However, the evaluation of their performance is still under discussion.…”