Covariance Biplot Analysis of Trace Element Concentrations in Urinary Stones

Abstract

The covariance biplot, a relatively new technique for displaying multivariate data, was applied to trace element contents and compound concentrations of urinary stones. The biplot is demonstrated to give a compact graphical representation of the multivariate data with interpretations in terms of familiar statistical concepts such as correlations and standard deviations. It displays strong correlations between various trace elements like Zn and Sr, and Sr and Na. The biplot also suggests concentration relationships which could play a hitherto unknown role in the genesis of calculi. It is shown to help in the interpretation of analytical results as well as in exposing erroneous or incomplete analyses.