Assessing Demographic Biases in Deep Learning Model for Fetal Growth Estimation in Clinical Practice. Patients Eligible for Inclusion Are Women With a Gestational Age Between 24-42 Weeks Undergoing a Third-trimester Growth Scan. The Image Data From the Scan Are Used to Calculate Fetal Weight.

Assessing Demographic Biases in Deep Learning Model for Fetal Growth Estimation in Clinical Practice

The goal of this observational study is to compare a new artificial intelligence (AI) feedback tool with the traditional method for estimating fetal weight during ultrasound scans on pregnant women between 24-42 weeks of gestation. The study aims to investigate the presence of demographic bias in the AI model. The demographic factors examined in the study include Body Mass Index (BMI), the number of births, fetal age, mother's age, fetal sex, and the presence of preeclampsia. Moreover, the study will compare the accuracy of the AI model and the Hadlock model, a fetal growth formula, in estimating fetal weight. Participants will have their ultrasound scans pseudonymized and securely stored on password-protected removable drives, ensuring their identity and privacy are maintained. Afterward, the ultrasound data will be sent to the Technical University of Denmark (DTU), where the AI model will analyze the images to estimate fetal weight.

Study Overview

• Status: Not yet recruiting
• Conditions: Pregnancy Complications

• Study Type: Observational
• Enrollment (Estimated): 300

Contacts and Locations

• Study Contact: Name: Julie Leth-Petersen; Phone Number: +4526399590; Email: julie.leth-petersen@regionh.dk
• Study Contact Backup: Name: Martin Grønnebæk Tolsgaard; Phone Number: 38 66 46 31; Email: martin.groennebaek.tolsgaard@regionh.dk

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Department of Prenatal Examinations at Rigshospitalet, Copenhagen, Denmark.

Description

Inclusion Criteria:

• Women with gestational age between 24-42 weeks undergoing a third-trimester growth scan.

Exclusion Criteria:

• Women with multiple pregnancies.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Pregnant women between 24-42 weeks of gestation; No interventions

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Demographic biases	The primary objective is to investigate potential demographic biases inherent in the deep learning model developed for estimating fetal growth in clinical practice. This is achieved by comparing the relative error between fetal weight at scan time (this value is extrapolated from the birth weight using the Marsal growth curve) and estimations from the Hadlock formula and the deep learning model.	From enrollment to the birth of the child

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Comparing the accuracy of the Hadlock formula and the AI model	The secondary objective is to compare the accuracy of fetal weight estimation between the Hadlock formula and a deep learning model in clinical practice using a paired t-test.	From enrollment to the birth of the child

Collaborators and Investigators

• Sponsor: Copenhagen Academy for Medical Education and Simulation

Publications and helpful links

General Publications

• Salomon LJ, Alfirevic Z, Da Silva Costa F, Deter RL, Figueras F, Ghi T, Glanc P, Khalil A, Lee W, Napolitano R, Papageorghiou A, Sotiriadis A, Stirnemann J, Toi A, Yeo G. ISUOG Practice Guidelines: ultrasound assessment of fetal biometry and growth. Ultrasound Obstet Gynecol. 2019 Jun;53(6):715-723. doi: 10.1002/uog.20272.

Study record dates

Study Major Dates

• Study Start (Estimated): March 15, 2024
• Primary Completion (Estimated): July 30, 2024
• Study Completion (Estimated): July 30, 2024

Study Registration Dates

• First Submitted: March 6, 2024
• First Submitted That Met QC Criteria: March 13, 2024
• First Posted (Actual): March 15, 2024

Study Record Updates

• Last Update Posted (Actual): March 15, 2024
• Last Update Submitted That Met QC Criteria: March 13, 2024
• Last Verified: March 1, 2024

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Pregnancy Complications
• Other Study ID Numbers: p-2024-15469

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No