cvd_bayes_net: CVD risk synthetic dataset

In laderast/cvdRiskData: Synthetic Dataset for Cardiovascular Risk Prediction

Description

A bayesian network for generating categorical synthetic data for assessing cardiovascular risk. Variable types are as follows:

Usage

data(cvd_bayes_net)

Format

A Bayesian Network of class CPTgrain using the gRain package for representing the data.

Details

• age Patient Age Category. Age category of patient. string of age ranges.

• htn Does patient have hypertension? Threshold systolic blood pressure is 150. Y/N

• treat Is patient receiving hypertension treatment? Y/N

• smoking Y/N based on threshold of pack years: 10

• race Race based on self-defined question in survey. AmInd (american indian), Asian/PI (asian/pacific islander), Black/AfAm (Black/African American), White

• gender Gender of patient. Male, Female, NA means that patient did not want gender recorded.

• t2d Whether patient has Type 2 diabetes. Y/N.

• bmi Body Mass Index of Patient. kg/m^2

• sbp Systolic Blood Pressure in mm/Hg

• rs10757278 SNP data. Associated with race and total cholesterol.

• rs1333049 SNP data. Associated with race and total cholesterol. Always co-occurs with rs10757278.

• rs4665058 SNP data. Associated with race and total cholesterol.

• rs8055236 SNP data. Variant is associated with increased risk.

Note that not all covariates (including cardiovascular risk) were generated by the dataset. Further details about how to generate the entire dataset from this network can be found from generate_data_from_network in the vignettes folder.

Examples

library(gRain)
data(cvd_bayes_net)

#generate categorical data for 1000 patients
testData <- simulate(cvd_bayes_net, nsim =1000)
summary(testData)

laderast/cvdRiskData documentation built on May 28, 2019, 7:37 a.m.