Making the Most Use of Data in Order to Provide Accurate Clinical Decision Support Systems (CDSSs) for Use in the Determination of Heart Disease: The Burn-Thornton Curves

摘要

In this paper we present curves, derived from experimental investigations, which will enable physicians to determine the minimum size of data set required to train a clinical decision support system, for use in the accurate determination of the presence of heart disease. The minimum size of data set required is a function of stratification bias.