Fetal Weight

Purpose: The primary objective of this study is to assess the diagnostic accuracy of an AI-enabled ultrasound tool for estimating fetal weight Participants: 1,000 pregnant individuals Procedures (methods): This prospective diagnostic accuracy study will enroll 1,000 pregnant individuals within one week of anticipated delivery. At a single visit, each participant will undergo two ultrasound assessments: (1) standardized sweeps for AI analysis (performed by both specialist and nonspecialist users), (2) specialist-performed fetal biometry.

Obstetric ultrasonography is now an essential tool for monitoring pregnancy. In France, three ultrasounds are recommended during a singleton pregnancy (at 12, 22, and 32 weeks of amenorrhea) to improve maternal and neonatal care. Among the parameters studied, fetal weight estimation allows for the assessment of fetal growth and the detection of certain conditions such as intrauterine growth restriction or macrosomia. Fetal weight estimation is based primarily on the Hadlock formula, which combines several biometric measurements (head circumference, abdominal circumference, and femur length). This estimate has a margin of error of 6 to 10% compared to birth weight, which is still accurate enough to guide important medical decisions, such as inducing labor or performing a cesarean section. However, an estimation error can have negative consequences for both mother and child. Fetal macrosomia is defined as a birth weight greater than or equal to 4000 g. It affects approximately 5 to 10% of pregnancies. Screening is mainly based on ultrasound, particularly in the second and third trimesters. Macrosomia is associated with an increased risk of complications for the mother (cesarean section, postpartum hemorrhage, or deep vein thrombosis) and the child (shoulder dystocia, fractures, brachial plexus palsy, or neonatal asphyxia). These risks are increased in cases of macrosomia in the context of maternal diabetes. Within the gynecology-obstetrics department, physicians use the protocole of macrosomia screening described in the DAME trial. If macrosomia is suspected an additional ultrasound is performed at around 36 weeks of amenorrhea. The Hadlock formula is used to estimate fetal weight. The ultrasound criteria for suspecting macrosomia are an estimated fetal weight above the 95th percentile (90th percentile in cases of maternal diabetes) according to WHO curves. In these cases, induction is offered between 38 and 39 weeks of amenorrhea, if the cervix is favorable. A cesarean section is offered to the patient if the estimated fetal weight is greater than 5000 g in the absence of diabetes and greater than 4500 g in cases of associated diabetes. Nevertheless, the probability of a child being born macrosomic after ultrasound suspicion is between 17% and 80% on average, and 53% in our center. Although beneficial in the context of macrosomia, the effects of induced labor or cesarean delivery are not insignificant for the mother and her baby. In this context, the value of this study is to demonstrate that performing an ultrasound as close to the birth as possible could allow for a more reliable estimation of fetal weight and better detection of macrosomia in order to avoid unnecessary procedures. The investigators hypothesize that fetal weight estimation is more accurate when performed within 24 hours prior to delivery. When the assessment is conducted earlier, it inherently assumes a constant fetal growth rate until birth, which may not reflect actual growth patterns. A more precise estimation of birth weight could improve clinical decision-making and optimize maternal and neonatal care, potentially reducing unnecessary interventions such as labor induction or cesarean delivery.

Objective: To evaluate the impact of real-time AI feedback on fetal biometry accuracy and investigate the Expertise Reversal Effect-whether AI benefits diminish as user experience increases. Design: A stratified randomized trial of 75 participants (25 Novices, 25 Intermediates, 25 Experts). Users are randomized 1:1 to either AI-assisted or manual measurement groups. Outcomes: * Primary: EFW accuracy (MAPE) compared to actual birthweight. * Secondary: Procedure time, image quality, error relative to baseline scans, and cognitive workload (NASA-TLX).

In this project, the investigators aim to study how all these factors determine the cardiovascular status of a total of 1,800 mothers, 3 to 6 years after delivery. In addition, the investigators want to assess whether lifestyle and living conditions after childbirth may improve or worsen this imprint, since women often prioritize their families over themselves, making it more difficult to maintain a healthy lifestyle that could reduce their cardiovascular risk. Furthermore, the investigators will evaluate how environmental exposures influence their health, as well as explore potential strategies for prediction and prevention. The goal is to develop an easy-to-use algorithm or test that allows women and their physicians to assess this risk, ideally in the form of a mobile app. Although predictive algorithms for cardiovascular health already exist, most have been developed using predominantly male or older populations, and none have taken into account pregnancy-related events or environmental exposure - both of which are key determinants of women's cardiovascular health.

Fetal Abdominal subcutaneous tissue thickness (FASTT) can be easily measured during the routine ultrasound examination of pregnant women. Numerous reports have shown FASTT measurement to be a good way of evaluating subcutaneous fat tissue. However, to the best of our knowledge, no studies have investigated the association of FASTT with abnormal fetal growth in nondiabetic. For this reason, in this study we evaluated whether FASTT can predict birth weight or diagnose LGA and/or LBW infants in the third trimester.