Title: Clinical pathway clustering using graph b-colouring and Markov models

Authors: Haytham Elghazel, Veronique Deslandres, Kassem Kallel, Alain Dussauchoy

Addresses: Universite de Lyon, Universite Lyon 1, EA4125 LIESP, Villeurbanne, F-69622, France. ' Universite de Lyon, Universite Lyon 1, EA4125 LIESP, Villeurbanne, F-69622, France. ' Universite de Lyon, Universite Lyon 1, EA4125 LIESP, Villeurbanne, F-69622, France. ' Universite de Lyon, Universite Lyon 1, EA4125 LIESP, Villeurbanne, F-69622, France

Abstract: In this paper, a framework for clustering clinical pathways is proposed which is based on a hybrid model that uses the recent b-colouring clustering approach coupled with Markov chain models. The approach allows firstly associating each cluster with a set of dominant sequences which guarantee robust partitioning and gives a solution to the cluster representation problem. Then, each cluster is governed by a finite-state Markov chain model. These models can be used for predicting possible paths when a new patient is admitted for example, in order to help medical professionals to plan resources for the clinical process.

Keywords: sequence clustering; graph b-colouring; Markov models; cluster representation; clinical pathways; PMSI; resource planning; new patients; healthcare; diagnosis related groups; medical information systems.

DOI: 10.1504/IJBET.2010.029656

International Journal of Biomedical Engineering and Technology, 2010 Vol.3 No.1/2, pp.156 - 172

Published online: 30 Nov 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article