Title: Foetal weight estimation with descriptive statistics and correlation analysis of significant ultrasonographic parameter and fuzzy artmap classifier

Authors: Saba Izadi; Somayeh Saraf Esmaili

Addresses: Department of Biomedical Engineering, Islamic Azad University, South Tehran Branch, Tehran, Iran ' Department of Biomedical Engineering, Islamic Azad University, Garmsar Branch, Garmsar, Iran

Abstract: High and low weight of the foetus at birth can be associated with an increased risk of neonatal complications. So far, various techniques have been proposed for estimating birth weight. In the proposed method a powerful fuzzy-neural classifier is used. The method is evaluated on a set of 40 ultrasonographic foetal data, in which foetuses are at 37 and 38 weeks of gestation. The features used for classification training and testing are superior features that have been used by experts in the field for many years, including the length of the femur, the bicuspid diameter, and the circumference of the foetal head. The results of the implementation of the proposed method on the dataset indicate the achievement of 98.96% accuracy, which will be evidence of its good performance on the new data. The new method can be used to provide accurate estimates of foetal birth weight.

Keywords: foetal weight estimation; ultrasonographic parameter; statistics and correlation analysis; fuzzy artmap - ANFIS classifier.

DOI: 10.1504/IJBRA.2023.131277

International Journal of Bioinformatics Research and Applications, 2023 Vol.19 No.1, pp.43 - 55

Received: 19 Nov 2022
Accepted: 08 Feb 2023
Published online: 05 Jun 2023 *