Skip to main content
. Author manuscript; available in PMC: 2014 Jun 1.
Published in final edited form as: J Biomed Inform. 2013 Apr 4;46(3):480–496. doi: 10.1016/j.jbi.2013.03.008

Table 3.

Feature description for each clinical dataset, where numerical covariates are indicated with “*” and categorical covariates are converted into binary covariates through dummy coding.

Dataset 7 Dataset 8 Dataset 9 Dataset 10 Dataset 11
x1 CA19* Birth number* Age* Symptoms
  1. Strongly agree

  2. Agree

  3. Disagree

  4. Strongly disagree

Pain in right arm
x2 CA125* Smoking status Beck Depression Score* Nausea
x3 - Race
  1. white

  2. black

  3. other

Drug use history
  1. Never

  2. Previous

  3. Recent

Hypo perfusion
x4 - Perceived benefit* ST elevation
x5 - Age of mother* Prior drug treatments* Family history New Q waves
x6 - Weight at last menstrual period* Race Self examination ST depression
x7 - - Treatment assignment Detection probability
  1. Not likely

  2. Somewhat likely

  3. Very likely

T wave inversion
x8 - - Treatment site Sweating
x9 - - - - Pain in left arm
y Presence of cancer Low birth weight Remained drug free Mammograph experience Presence of disease
HHS Vulnerability Disclosure