Efficacy of interferon treatment for chronic hepatitis C predicted by feature subset selection and support vector machine

J Med Syst. 2007 Apr;31(2):117-23. doi: 10.1007/s10916-006-9046-8.


Chronic hepatitis C is a disease that is difficult to treat. At present, interferon might be the only drug, which can cure this kind of disease, but its efficacy is limited and patients face the risk of side effects and high expense, so doctors considering interferon must make a serious choice. The purpose of this study is to establish a simple model and use the clinical data to predict the interferon efficacy. This model is a combination of Feature Subset Selection and the Classifier using a Support Vector Machine (SVM). The study indicates that when five features have been selected, the identification by the SVM is as follows: the identification rate for the effective group is 85%, and the ineffective group 83%. Analysis of selected features show that HCV-RNA level, hepatobiopsy, HCV genotype, ALP and CHE are the most significant features. The results thus serve for the doctors' reference when they make decisions regarding interferon treatment.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Antiviral Agents / therapeutic use*
  • Biopsy
  • Decision Support Systems, Clinical / organization & administration*
  • Female
  • Genotype
  • Hepatitis C, Chronic / blood
  • Hepatitis C, Chronic / drug therapy*
  • Humans
  • Interferons / therapeutic use*
  • Liver Function Tests
  • Male
  • Middle Aged
  • RNA / blood


  • Antiviral Agents
  • RNA
  • Interferons