Advanced Prenatal Care With AI-Enabled Ultrasound to Estimate Gestational Age

The ability to accurately determine gestational age (GA) is crucial for optimal prenatal care. However, traditional ultrasonography, which is necessary for precise GA estimation, may not be available in low-resource settings. A recent study has shown promising results for an innovative solution: a low-cost, battery-powered ultrasonography probe integrated with artificial intelligence (AI) for image interpretation. This tool allows novice clinicians to estimate GA with remarkable accuracy, comparable to credentialed sonographers using conventional ultrasonography devices (1^✔ ✔Trusted Source
Diagnostic Accuracy of an Integrated AI Tool to Estimate Gestational Age From Blind Ultrasound Sweeps

Go to source). Accurate assessment of gestational age is essential for good pregnancy care. In low-resource settings, access to conventional ultrasonography is often limited, posing a significant challenge. This new study explores an AI-enabled ultrasonography device that allows novice users to estimate GA with precision comparable to traditional methods.

This prospective diagnostic accuracy study enrolled 400 pregnant women with viable, single, non-anomalous first-trimester pregnancies. The study was conducted in Lusaka, Zambia, and Chapel Hill, North Carolina. Credentialed sonographers established the baseline GA using transvaginal crown-rump length measurement. During follow-up visits scheduled randomly throughout gestation, novice clinicians performed blind sweeps of the maternal abdomen using the AI-enabled device, while credentialed sonographers conducted fetal biometry using high-specification machines.

Accuracy and Consistency of AI-Enabled Device to Determine Gestational Age

The primary outcome measured was the mean absolute error (MAE) of the AI-enabled device compared to the traditional ultrasonography device. The results demonstrated that the AI tool was equivalent to the traditional method, with an MAE of 3.2 days compared to 3.0 days (difference of 0.2 days). Furthermore, 90.7% of assessments with the AI device were within 7 days of the established GA, compared to 92.5% for the traditional method.

The AI tool’s performance was consistent across various subgroups, including those with higher body mass index (BMI) and participants from both Zambia and North Carolina. This consistency highlights the tool’s robustness and adaptability to different populations and conditions.

Health Implications in Low-Resource Settings and Improved Maternal and Fetal Outcomes

The study’s findings suggest that novice users, even without prior ultrasonography training, can accurately estimate GA using the AI-enabled device. This breakthrough has significant implications for improving obstetrical care in low-resource settings, aligning with the World Health Organization’s goal of providing GA estimation for all pregnant individuals.

The integration of AI in ultrasonography has the potential to revolutionize prenatal care, especially in low-resource settings. By enabling novice clinicians to accurately estimate GA, this technology can ensure timely and appropriate prenatal interventions, ultimately improving maternal and fetal outcomes.

Advertisement

Between 14 and 27 weeks of gestation, novice users with no prior ultrasonography training were able to estimate GA with the low-cost, AI-enabled device as accurately as credentialed sonographers using high-specification machines. This study demonstrates the potential of AI integration in ultrasonography to enhance obstetrical care in low-resource settings, paving the way for broader accessibility to essential prenatal diagnostics.

Reference:

1. Diagnostic Accuracy of an Integrated AI Tool to Estimate Gestational Age From Blind Ultrasound Sweeps - (https://jamanetwork.com/journals/jama/article-abstract/2821666)

Source-Medindia