One of the latest advancements in ultrasound is the incorporation of artificial intelligence (AI). AI-assisted ultrasound imaging applications and tools are being developed to aid medical practitioners in their decision-making in gynecology and obstetrics. The goal is to address longstanding challenges in diagnosis and treatment, ultimately supporting and improving maternal and fetal health.
To illustrate how AI can enhance detection rates of malformations and fetal diagnosis, a paper in the UOG journal by Taksoee-Vester et al. (2024) demonstrates the potential impact. Implementing AI can ultimately lead to improved outcomes for infants with Coarctation of the Aorta (CoA) by enabling early intervention and treatment.
Congenital heart disease causes 30% of infant deaths from congenital malformations, with coarctation of the aorta (CoA) representing 5-8% of these cases.
The paper reveals that around 60% of newborns with isolated CoA are not diagnosed before birth. Prenatal detection of CoA greatly impacts survival rates, emphasizing how important it is for better detection methods.
The research team set out to identify reliable quantitative fetal echocardiographic predictors that could predict the postnatal development of Coarctation of the Aorta (CoA). Their goal was to develop an AI screening tool that could be used during the 18-22-week scan. They believed that using automated cardiac measurements with AI could improve detection accuracy for CoA.
The study results confirmed their hypothesis that during the 18-22 week scan, significant differences were observed in the right ventricle (RV) area and length, left ventricle (LV) width, and the ratios of RV/LV areas and main pulmonary artery/ascending aorta diameters with z-scores above 0.7 when comparing subjects with a postnatal diagnosis of Coarctation of the Aorta (CoA) (n=73) and healthy controls (n=7300).
The information tells us that the cardiac biometric measurements show significant differences between fetuses later diagnosed with Coarctation of the Aorta (CoA) and healthy controls during the 18-22 week scan.
Therefore, the study is critical in telling us that AI should be leveraged as a standardised method for conducting cardiac biometric measurement, which can enhance the detection and early intervention of CoA.
Please click here to view all of the 2024 issues of the UOG Journal and to view the full article here. To view all UOG content, become an ISUOG member today or login and upgrade.
Reference
Taksoee-Vester A, Jørgensen AS, Petersen OB, et al. Automated cardiac biometric measurements with artificial intelligence for prenatal detection of coarctation of the aorta. Ultrasound in Obstetrics & Gynecology. 2024;54(3):345-352. doi:10.1002/uog.27608