Sources:
    (a) Original Owner:
          Danny Coomans, Department of Maths. and Stats., James Cook University, Townsville 4811, Australia.
          email:  madhc@manta.jcu.edu.au
    (b) Donor of database
          Stefan Aeberhard, Department of Comp. Science, James Cook University, Townsville 4811, Australia.
          email:  stefan@coral.cs.jcu.edu.au
    (c) Date Received
          July, 1992

  Past Usage:
    (a) Coomans, D., Broeckaert, M. Jonckheer M. and Massart D. L.,
          "Comparison of Multivariate Discriminant Techniques for Clinical Data - Application to the Thyroid Functional State,"
          Methods of Information in Medicine, Vol. 22, 1983, pp. 93-101.
    (b) Coomans, D. and I. Broeckaert,
          "Potential Pattern Recognition in Cemical and Medical Decision Making,"
          Research Studies Press, Letchworth, England.

  Relevant Information Paragraph:
    Five laboratory tests are used to try to predict whether a patient's thyroid to the class euthyroidism, hypothyroidism or hyperthyroidism.
    The diagnosis (the class label) was based on a complete medical record, including anamnesis, scan etc.

  Number of Instances:  215
  Number of Features:   5
  Class Distribution:
      Class 1: (Normal)  150
      Class 2: (Hyper)    35
      Class 3: (Hypo)     30

  Features:
      1:   Class label. (1 = normal, 2 = hyper, 3 = hypo)
      2:   T3-resin uptake test. (%)
      3:   Total Serum thyroxin as measured by the isotopic displacement method.
      4:   Total Serum triiodothyronine as measured by radioimmuno assay.
      5:   Basal thyroid-stimulating hormone (TSH) as measured by radioimmuno assay.
      6:   Maximal absolute difference of TSH value after injection of 200 micro grams of thyrotropin-releasing hormone as compared to the basal value.