Advertisement

Classification Method for Differential Diagnosis Based on the Course of Episode of Care

  • Adrian Popiel
  • Tomasz Kajdanowicz
  • Przemyslaw Kazienko
  • Jean Karl Soler
  • Derek Corrigan
  • Vasa Curcin
  • Roxana Danger Mercaderes
  • Brendan Delaney
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8073)

Abstract

The main goal of the paper is to propose a classification method for differential diagnosis in primary care domain. Commonly, the final diagnosis for the episode of care is related with the initial reason for encounter (RfE). However, many distinct diagnoses can follow from a single RfE and they need to be distinguished. The new method exploits the data about whole episodes of care quantified by individual patients’ encounters and it extracts episode features from electronic health record to learn the classifier. The experimental studies carried out on two primary care dataset from Malta and the Netherlands for three distinct diagnostic groups revealed the validity of the proposed approach.

Keywords

Classification Differential Diagnosis Classification Episode of Care Diagnosis 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Soler, J.K., Okkes, I.: Reasons for encounter and symptom diagnoses: a superior description of patients’ problems in contrast to medically unexplained symptoms (mus). Family Practice 29(3), 272–282 (2012)CrossRefGoogle Scholar
  2. 2.
    Schiff, G., Bates, D.: Can electronic clinical documentation help prevent diagnostic errors? The New England Journal of Medicine 362(12), 1066–1069 (2010)CrossRefGoogle Scholar
  3. 3.
    Romano, M., Stafford, R.: Electronic health records and clinical decision support systems. Arch. Intern. Med. 171(10), 897–903 (2011)CrossRefGoogle Scholar
  4. 4.
    Kortteisto, T., Komulainen, J., Kunnamo, I., Makela, M., Kaila, M.: Implementing clinical decision support for primary care professionals – the process. Finnish Journal of eHealth and eWelfare 4(3), 153–164 (2012)Google Scholar
  5. 5.
    Yoo, I., Alafaireet, P., Marinov, M.: Data mining in healthcare and biomedicine: A survey of the literature. J. Med. Syst. 36, 2431–2448 (2012)CrossRefGoogle Scholar
  6. 6.
    Li, J., Fu, A., Fahey, P.: Efficient discovery of risk patterns in medical data. Artificial Intelligence in Medicine 45, 77–89 (2009)CrossRefGoogle Scholar
  7. 7.
    Yang, F., Wang, H.: Using random forest for reliable classification and cost-sensitive learning for medical diagnosis. In: The Seventh Asia Pacific Bioinformatics Conference (2009)Google Scholar
  8. 8.
    WONCA: An introduction to the international classification of primary care version 2. Technical report, World Organization of Family Doctors (WONCA), WONCA International Classification Committee (WICC) (2004)Google Scholar
  9. 9.
    Olson, D., Delen, D.: Advanced Data Mining Techniques. Springer (2008)Google Scholar
  10. 10.
    Tiwari, A., Sekhar, A.K.T.: Review article: Workflow based framework for life science informatics. Comput. Biol. Chem. 31(5-6), 305–319 (2007)zbMATHCrossRefGoogle Scholar
  11. 11.
    Frank, E., Hall, M.A., Holmes, G., Kirkby, R., Pfahringer, B., Witten, I.H., Trigg, L.: Weka - a machine learning workbench for data mining. In: Maimon, O., Rokach, L. (eds.) The Data Mining and Knowledge Discovery Handbook, pp. 1305–1314. Springer (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Adrian Popiel
    • 1
  • Tomasz Kajdanowicz
    • 1
  • Przemyslaw Kazienko
    • 1
  • Jean Karl Soler
    • 2
  • Derek Corrigan
    • 3
  • Vasa Curcin
    • 4
  • Roxana Danger Mercaderes
    • 4
  • Brendan Delaney
    • 5
  1. 1.Institute of InformaticsWroclaw University of TechnologyWroclawPoland
  2. 2.Mediterranean Institute of Primary CareAttardMalta
  3. 3.Department of General PracticeRoyal College of Surgeons in IrelandIreland
  4. 4.Department of ComputingImperial College LondonUnited Kingdom
  5. 5.Kings College LondonUnited Kingdom

Personalised recommendations