Towards evolutionary discovery of typical clinical pathways in electronic health records

Abstract According to value-based health care, hospitals should deliver high quality and cost-effective medical treatments. Clinical pathways represent typical ways of treatment and indicate its impact on hospital resources. In this paper, we propose three approaches to discover and to cluster clinical pathways using the k-means method and genetic algorithms. A data set of 3434 electronic health records of patients with acute coronary syndrome is used to test proposed approaches. The approaches are compared with five clustering metrics. The best approach shows the best values for four of five metrics. In the future, we plan to embed the best approach of discovering clinical pathways in personalized Decision Support System of Almazov National Medical Research Centre (Saint Petersburg, Russia).

[1]  Siddheswar Ray,et al.  Determination of Number of Clusters in K-Means Clustering and Application in Colour Image Segmentation , 2000 .

[2]  Sergey V. Kovalchuk,et al.  Simulation of Patients Flow in Healthcare Multiple Units using Process and Data Mining Techniques for Model Identification , 2017 .

[3]  David E. Goldberg,et al.  A niched Pareto genetic algorithm for multiobjective optimization , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

[4]  Gary C. Brown,et al.  Update on Value-Based Medicine , 2013, Current opinion in ophthalmology.

[5]  Alexey N. Yakovlev,et al.  Complex Data-driven Predictive Modeling in Personalized Clinical Decision Support for Acute Coronary Syndrome Episodes , 2016, ICCS.

[6]  Jong-Myon Bae,et al.  Value-based medicine: concepts and application , 2015, Epidemiology and health.

[7]  Huilong Duan,et al.  Summarizing clinical pathways from event logs , 2013, J. Biomed. Informatics.

[8]  Wil M. P. van der Aalst,et al.  Fuzzy Mining - Adaptive Process Simplification Based on Multi-perspective Metrics , 2007, BPM.

[9]  Dirk Fahland,et al.  Conformance Checking Based on Partially Ordered Event Data , 2014, Business Process Management Workshops.

[10]  Jason Brownlee,et al.  Clever Algorithms: Nature-Inspired Programming Recipes , 2012 .

[11]  E. James,et al.  What is a clinical pathway? Development of a definition to inform the debate , 2010, BMC medicine.

[12]  Wil M. P. van der Aalst,et al.  Genetic Process Mining: A Basic Approach and Its Challenges , 2005, Business Process Management Workshops.

[13]  Jong-Myon Bae,et al.  Development and application of patient decision aids , 2015, Epidemiology and health.

[14]  Xiaolan Xie,et al.  Evaluation of discovered clinical pathways using process mining and joint agent-based discrete-event simulation , 2016, 2016 Winter Simulation Conference (WSC).

[15]  Ian F. C. Smith,et al.  A Bounded Index for Cluster Validity , 2007, MLDM.

[16]  Fei Wang,et al.  Mining and exploring care pathways from electronic medical records with visual analytics , 2015, J. Biomed. Informatics.

[17]  Liz Sonenberg,et al.  MINING SURPRISING PATTERNS AND THEIR EXPLANATIONS IN CLINICAL DATA , 2014, Appl. Artif. Intell..

[18]  Gary C. Brown,et al.  Health care economic analyses and value-based medicine. , 2003, Survey of ophthalmology.

[19]  Wil M. P. van der Aalst,et al.  Workflow mining: discovering process models from event logs , 2004, IEEE Transactions on Knowledge and Data Engineering.

[20]  Rui Liu,et al.  Pathway-Finder: An Interactive Recommender System for Supporting Personalized Care Pathways , 2014, 2014 IEEE International Conference on Data Mining Workshop.

[21]  Richard Herrmann,et al.  IGF-I induced genes in stromal fibroblasts predict the clinical outcome of breast and lung cancer patients , 2010, BMC medicine.

[22]  Wil M. P. van der Aalst,et al.  Analyzing “Spaghetti Processes” , 2011 .

[23]  Catherine Burns,et al.  An Exploratory Case Study to Understand Primary Care Users and Their Data Quality Tradeoffs , 2017, ACM J. Data Inf. Qual..

[24]  Vilas M. Thakare,et al.  Computing the Most Significant Solution from Pareto Front obtained in Multi-objective Evolutionary , 2010 .