Simulation and critical care modeling

Abstract

Simulation models provide useful tools for organizing and analyzing the interactions between therapies, tradeoffs, and outcomes.

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.…”

Health Economic Methods: Cost-Minimization, Cost-Effectiveness, Cost-Utility, and Cost-Benefit Evaluations

“…30 Markov models require development of a series of states that must be mutually exclusive (a patient can be in only one state at a time) and collectively exhaustive (the patient has to be in one of the states at all times). 31 Patients move through the model according to probabilities that determine how likely it is to transition from one state to another over a specified period of time (the Markov cycle). Each health state is associated with a certain cost and health outcome for each cycle.…”

“…To run the Markov model, individual patient cohorts cycle through the different states of the model based on the transition probabilities. 31 The model is run either for a fixed number of cycles or until all patients have entered an absorbing state (a state that they will remain in for all future cycles). 31 A simple characterization of a Markov model is shown in Fig.…”

Prediction of survival probabilities with Bayesian Decision Trees

“…The authors of review [30] found that conventional modelling techniques used in critical care could be improved by simulation methods such as MCMC. They analysed the usefulness and limitations of applications of these methods presented in the related literature and concluded that simulation provided practitioners with additional information on risks.…”