Ethnicity as a factor for the estimation of the risk for preeclampsia: A neural network approach

Abstract

A large number of feedforward neural structures, both standard multilayer and multi-slab schemes have been applied to a large data base of pregnant women, aiming at generating a predictor for the risk of preeclampsia occurrence at an early stage. In this study we have investigated the importance of ethnicity on the classification yield. The database was composed of 6838 cases of pregnant women in UK, provided by the Harris Birthright Research Centre for Fetal Medicine in London. For each subject 15 parameters were considered as the most influential at characterizing the risk of preeclampsia occurrence, including information on ethnicity. The same data were applied to the same neural architecture, after excluding the information on ethnicity, in order to study its importance on the correct classification yield. It has been found that the inclusion of information on ethnicity, deteriorates the prediction yield in the training and test (guidance) data sets but not in the totally unknown verification data set.