Create a Mathematical Model to Estimate the Weight of Iraqi Fetuses Using Ultrasound Images

1Ahlam A. Hussain and Ebtesam F. Kanger


Background: Ultrasound (US) imaging is a medical imaging technique, considered one of the most used techniques in hospitals and clinical centers around the world, due to it is safe, economical, transferable, and adaptable. US imaging using to diagnosis human body parts and their functions, the most common using of US in pregnancy to monitor the developments of fetuses and revealing of abnormalities, estimation gestational age (GA), as well as estimation fetal weight (EFW). EFW is an important component of maternity health care, especially in the management and planning of the delivery mode, this will play a major role in newborn and maternal safety, EFW depend of measurement the fetal biometry such as abdominal circumference (AC), femur length (FL), bi-partial diameter (BPD) and others, accurate measurement of fetal biometry depend on the experience of sonographers. There are several models were used to EFW depending on one or more biometric of fetuses, but these models still given a some of errors in EFW in Iraq. So, this study has been carried out for creating the first EFW model for the first time in Iraq and the Middle East, a study is conducted to determine the weight of the fetus by finding a special mathematical model for the Iraqi fetuses. Depending on the US images taken from Iraqi hospitals, as will be explained in this search. Methods: This study was performed in the department of Obstetrics and Gynecology in Al- Yarmouk Teaching Hospital and AL- Alawiya Teaching Hospital in Baghdad, Iraq, during 2019 on 200 pregnant women of singleton and normal pregnancies, fetal GA (36-40) week (W), and the last US scan was less than 7 days before birth. The obtained dataset (fetal biometry and actual birth weight (ABW)), were utilized to developed EFW using IBM SPSS Version 23 software package ( Software). Results: Statistical analysis indicated that there was a significant difference between the ABW and EFW using most weight estimation models, only our proposed model and Hadlock2 model, produced the acceptable results, where the correlation between ABW and EFW by proposed model was (R=0.964), by Hadlock2 model (R=0.920), while it extremely declined when using other models and showed less correlation. By using Hsieh model was (R=0.330), by Woo model (R=0.236), and by Jordaan model (R=0.667). Conclusion: The proposed model and Hadlock2 model produce results better than other models in EFW for Iraqi pregnant, but depending on the absolute mean error (AME), absolute percentage error (APE), and R value, we conclude that, the proposed model was the best model for EFW in Iraq, and gave most acceptable results compared with ABW.


Ultrasound Imaging, Pregnancy, Fetal Biometry, Estimated Fetal Weight Models.

Paper Details
IssueIssue 5