A Process Mining Driven Framework for ClinicalGuideline Improvement in Critical Care

Learning from Medical Data Streams 2011.

医药卫生;计算机科学

Carolyn McGregor1 ; Christina Catley1and Andrew James2 ; 3

Others  :  http://ceur-ws.org/Vol-765/paper5.pdf
PID  :  41148

来源: CEUR

【 摘 要 】

This paper presents a framework for process mining in critical care. The framework uses the CRISP-DM model, extended to incorporate temporaland multidimensional aspects (CRISP-TDMn), combined with the PatientJourney Modeling Architecture (PaJMa), to provide a structured approach toknowledge discovery of new condition onset pathophysiologies in physiologicaldata streams.The approach is based on temporal abstraction and mining ofphysiological data streams to develop process flow mappings that can be usedto update patient journeys; instantiated in critical care within clinical practiceguidelines. We demonstrate the framework within the neonatal intensive caresetting, where we are performing clinical research in relation topathophysiology within physiological streams of patients diagnosed with lateonset neonatal sepsis. We present an instantiation of the framework for lateonset neonatal sepsis, using CRISP-TDMn for the process mining model andPaJMa for the knowledge representation.