Simulation of Patient Flow in Multiple Healthcare Units using Process and Data Mining Techniques for Model Identification

摘要

Introduction: An approach to building a hybrid simulation of patient flow isintroduced with a combination of data-driven methods for automation of modelidentification. The approach is described with a conceptual framework and basicmethods for combination of different techniques. The implementation of theproposed approach for simulation of acute coronary syndrome (ACS) was developedand used within an experimental study. Methods: Combination of data, text, andprocess mining techniques and machine learning approaches for analysis ofelectronic health records (EHRs) with discrete-event simulation (DES) andqueueing theory for simulation of patient flow was proposed. The performedanalysis of EHRs for ACS patients enable identification of several classes ofclinical pathways (CPs) which were used to implement a more realisticsimulation of the patient flow. The developed solution was implemented usingPython libraries (SimPy, SciPy, and others). Results: The proposed approachenables more realistic and detailed simulation of the patient flow within agroup of related departments. Experimental study shows that the improvedsimulation of patient length of stay for ACS patient flow obtained from EHRs inFederal Almazov North-west Medical Research Centre in Saint Petersburg, Russia.Conclusion: The proposed approach, methods, and solutions provide a conceptual,methodological, and programming framework for implementation of simulation ofcomplex and diverse scenarios within a flow of patients for different purposes:decision making, training, management optimization, and others.