Fetal Birth Weight Estimation with Machine Learning Techniques in 15-40 Weeks of Pregnancy

Aug 4, 2026

Upload by Aria

Accurate prediction of fetal birth weight is essential for maternal and fetal health, since low birth weight (<2500 g) and high birth weight (>4000 g) increase perinatal mortality, complications, and long-term chronic disease risk. This study analyzes 730 singleton fetuses from 15–40 weeks using clinical variables and ultrasound biometry, applying nine supervised regression models to estimate birth weight and support clinicians in risk detection before delivery.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

Introduction

Materials and Methods

Results

Conclusion

Key words

Introduction (Background)

Significance of Birth Weight Prediction

Limitations of Direct Measurement and Traditional Estimation

Prior Ultrasound-Based Formulas and Modeling Approaches

Role of Neural Networks and Machine Learning

Study Rationale for the Karaman Region

Fetal Birth Weight Estimation with Machine Learning Techniques in 15-40 Weeks of Pregnancy Özlem Dülger1 , Ahmet Dursun2 , Usame Ömer Osmanoğlu3 1Karamanoğlu Mehmetbey University, Faculty of Medicine, Department of Obstetrics and Gyneacology, Karaman, Türkiye 2Karamanoğlu Mehmetbey University, Faculty of Medicine, Department of Anatomy, Karaman, Türkiye 3Karamanoğlu Mehmetbey University, Faculty of Medicine, Department of Biostatistics, Karaman, Türkiye Abstract Introduction: Accurate prediction of birth weight is crucial for both fetuses and mothers. Low birth weight (birth weight < 2500g) and high birth weight (birth weight > 4000g) can lead to high perinatal mortality rates, various complications, and both short-term and longterm health outcomes such as chronic diseases. This article aims to propose a machine-learning solution to enhance the accuracy of birth weight prediction and assist clinicians in identifying potential risks before birth. Materials and Methods: Seven hundred thirty different fetuses between weeks 15-40 were analyzed using clinical data. Nine different regression models from supervised machine learning methods, including logistic regression, support vector machine, decision t ree, elastic net regressor, lasso regressor, ridge regressor, artificial neural network, random forest, and k-nearest neighbors algorithms, were employed to predict fetal birth weights based on gestational week, maternal age, gender, and ultrasound measurements, including bipari etal diameter, abdominal circumference, and femur length. Results: Our study revealed that abdominal circumference was the most influential parameter, while gender had the least impact. The performance of the nine different algorithms in birth weight prediction was compared, and the elastic net regressor algorithm exhibited the best predictive performance. The proposed model yielded a prediction result with an average absolute error percentage of 8.87% and an average error of ±284g. A new formula for the newborn weight prediction model was developed using the elasti c net regressor machine learning method. Conclusion: Our study demonstrates that the model created with the elastic net regressor algorithm can predict birth weight at any gestational age between weeks 15-40. Key words: Pregnancy; birth weight; ultrasonography; artificial intelligence; machine learning. Introduction The well-being and survival of a fetus in utero are closely linked to intrauterine weight gain because fetuses outside the normal weight range are at increased risk of long-term perinatal morbidity, mortality, and poor growth and development (1) . The significance of accurately predicting birth weight is underscored by low birth weight being a risk factor for many adult diseases, reflecting a baby’s health in later life (2) . Accurate prediction of birth weight plays a crucial role in determining neonatal care requirements. Infants with low birth weight (<2500g) or high birth weight (>4000g) face higher risks of perinatal and postnatal complications compared to those with normal birth weight (2500-4000g) . Infants with low birth weight have significantly higher morbidity and mortality rates compared to those with normal birth weight. Potential risks associated with high birth weight include shoulder dystocia, intrapartum asphyxia, trauma, various maternal complications, and some metabolic complications (3) . However, birth weight cannot be directly measured before birth and is often roughly estimated based on clinicians’ experiences (4) . Researchers have attempted to predict fetal birth weight using single or multiple ultrasound measurement parameters (3) . The first successful approach involved the core correlation between fetal abdominal circumference measurement and birth weight (5) . Subsequently, researchers developed various formulas based on one, several, or all ultrasound parameters, such as abdominal circumference (AC), biparietal diameter (BPD), fe

FAQs

Why is fetal birth weight prediction important in pregnancy?

Birth weight outside the normal range is linked to higher perinatal morbidity and mortality and can influence neonatal care needs. Accurate prediction helps identify risks before delivery.

What data and ultrasound measurements are used to predict birth weight?

The study uses gestational week, maternal age, fetal gender, and ultrasound measurements including biparietal diameter, abdominal circumference, and femur length, with fetuses analyzed between weeks 15 and 40.

Which machine-learning model performed best and what accuracy was reported?

The elastic net regressor achieved the best predictive performance. The model reported an average absolute error percentage of 8.87% and an average error of ±284 g.

Aug 4, 2026

Upload by Aria