Neural networks to estimate the risk for preeclampsia occurrence

  • Authors:

  • Costas K. Neocleous;Panagiotis Anastasopoulos;Kypros H. Nikolaides;Christos N. Schizas;Kleanthis C. Neokleous

  • Affiliations:

  • Department of Mechanical Engineering, Cyprus University of Technology, Lemesos, Cyprus;Harris Birthright Research Centre for Fetal Medicine, King's College Hospital Medical School, London, United Kingdom;Harris Birthright Research Centre for Fetal Medicine, King's College Hospital Medical School, London, United Kingdom;Department of Computer Science, University of Cyprus, Nicosia, Cyprus;Department of Computer Science, University of Cyprus, Nicosia, Cyprus

  • Venue:
  • IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

A number of neural network schemes have been applied to a large data base of pregnant women, aiming at generating a predictor for the estimation of the risk of occurrence of preeclampsia at an early stage. 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, 24 parameters were measured or recorded. Out of these, 15 parameters were considered as the most influencing at characterizing the risk of preeclampsia occurrence. A number of feedforward neural structures, both standard multilayer and multi-slab, were tried for the prediction. The best results obtained were with a multi-slab neural structure. In the training set there was a correct classification of the 83.6% cases of preeclampsia and in the test set 93.8%. The preeclampsia cases prediction for the totally unknown verification test was 100%.