Artificial Intelligence-Driven Ultrasound Automation for the Early Prediction of Fetal Growth Restriction

International Journal of Obstetrics and Gynecology | Vol. 14, No. 5, May 2026 | pp. 1–12

DOI: 10.46882/2026/IJOG/000477

Original Research Article

Title: Artificial Intelligence-Driven Ultrasound Automation for the Early Prediction of Fetal Growth Restriction

Names of Authors: Chinedu O. Okafor¹, Fatima B. Aliyu²

Authors’ Affiliations:
¹ Department of Maternal-Fetal Medicine, University of Nigeria Teaching Hospital, Enugu, Nigeria
² Department of Obstetrics and Gynecology, Aminu Kano Teaching Hospital, Kano, Nigeria

Abstract:
Fetal Growth Restriction (FGR) remains a primary driver of stillbirth and neonatal morbidity. Traditional manual biometry presents notable inter-observer variability, potentially delaying diagnosis. This prospective multi-center study evaluated the diagnostic precision of an artificial intelligence (AI)-driven automated ultrasound system compared to standard manual biometry for the early identification of FGR. A cohort of 1,200 singleton pregnancies was evaluated during routine mid-trimester scans (18 to 22 weeks of gestation). The AI algorithm automatically segmented and measured the fetal abdominal circumference, head circumference, femur length, and mean uterine artery pulsatility index (UtA-PI). An ensemble machine learning model integrated these biometrics with maternal clinical risk factors to calculate an individualized FGR risk score. FGR was defined postnatally as a birth weight below the 10th percentile according to customized growth charts. The AI system achieved an automated biometry success rate of 98.2% on the first attempt, significantly reducing the average scan acquisition time from 14.5 ± 3.2 minutes to 4.1 ± 1.2 minutes (p < 0.001). For the prediction of early-onset FGR, the AI-integrated risk model demonstrated a sensitivity of 88.5%, a specificity of 94.1%, a positive predictive value of 68.2%, and a negative predictive value of 98.9%. In comparison, standard manual biometry coupled with traditional risk checklists yielded a sensitivity of 64.1% and a specificity of 87.3% (p < 0.001). In conclusion, deploying AI-driven ultrasound automation significantly enhances the early prediction of FGR, optimizes clinical workflows, and provides an objective, highly scalable screening method to improve perinatal surveillance.

Keywords: Fetal Growth Restriction; Artificial Intelligence; Automated Ultrasound; Deep Learning; Prenatal Screening; Perinatal Outcomes.

Manuscript Timeline: Received: January 10, 2026; Revised: March 04, 2026; Accepted: April 02, 2026; Published: May 05, 2026.

Citation: Okafor CO, Aliyu FB. Artificial Intelligence-Driven Ultrasound Automation for the Early Prediction of Fetal Growth Restriction. International Journal of Obstetrics and Gynecology. 2026; 14(5): 1–12.

Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read