[The importance of multivariate statistical procedures in the diagnosis of acute radiation injuries in man].

Abstract

Data published in international papers were analyzed and submitted to a multidimensional variate and discriminant analysis. In this analysis, information on prodromal symptoms and on hematological alterations 24 hours after irradiation was considered. As a result of the analysis it was pointed out that by means of a discriminant function the classification of persons to the four degrees of radiation sickness is possible with a computed false rate of classification of nearly 20% and a dose-effect-relation exists between discriminant scores and computed whole-body doses.