The Tohoku Journal of Experimental Medicine
Online ISSN : 1349-3329
Print ISSN : 0040-8727
ISSN-L : 0040-8727
Regular Contribution
Artificial Neural Network for Predicting Iodine Deficiency in the First Trimester of Pregnancy in Healthy Women
Maria Teresa Murillo-LlorenteCarmen Fajardo-MontañanaMarcelino Perez-Bermejo
ジャーナル オープンアクセス HTML

2020 年 252 巻 3 号 p. 185-191


Iodine deficiency in Spain is a persisting public health problem and the prescription of potassium iodide is recommended during pregnancy. The purpose of this study was to develop an Artificial Neural Network (ANN) to predict the risk factors of iodine deficiency during pregnancy, and compare the results obtained with a logistic regression model. Two hundred forty-four healthy pregnant women were included in a descriptive and prospective study in their first trimester of pregnancy. The women enrolled were asked specifically about their use of supplements containing potassium iodide, iron, folic acid and/or multivitamins during pregnancy. The consumption of iodine-rich foods was assessed through a food frequency questionnaire. A median UIC of 57.4 μg/L (IQR 32.8-99.3) was obtained, with 89.3% < 150 μg/L, the minimum recommended ioduria level by the WHO. There was no correlation between urinary iodine concentrations and maternal age, BMI or gestation week at recruitment. The urinary iodine concentrations were significantly higher in women who reported taking iodized supplements and/or iodized salt than those who did not. Number of gestations, age, body mass index, and intake of iodized supplements and iodized salt were the most important predictors of iodine deficiency. Based on Receiver Operating Characteristic analysis, the diagnostic performance of the ANN model was superior to the logistic regression model. The ANN model, with variables on pregnancy and the intake of iodine rich foods, iodized supplement and iodized salt may be useful for predicting iodine deficiency in the early pregnancy.

© 2020 Tohoku University Medical Press

This article is licensed under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license.
This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC-BY-NC-ND 4.0). Anyone may download, reuse, copy, reprint, or distribute the article without modifications or adaptations for non-profit purposes if they cite the original authors and source properly.