Untitled

Abstract

In fetal medicine, artificial intelligence plays a crucial role in preventing congenital fetal abnormalities. Anomalies of heart and brain in fetal ultrasonography and MRI have been shown to be recognizable, detectable, and localizable by ML algorithms and CNNs. Artificial Intelligence (AI) systems are capable of carrying out intricate analyses of aberrant image patterns in order to categorize and predict malformations in fetuses. The role of Artificial Intelligence (AI) in the prediction and risk stratification of congenital anomalies is explored in this narrative review. Fetal imaging (ultrasonography and MRI) examination may be optimized by ML and DL algorithms to reduce examination time, lighten the doctor's workload, and increase diagnostic precision for fetal anomalies. The current study's objective is to evaluate the algorithms being utilized to automate screening for fetal brain and heart anomalies. It also compares ML and DL algorithms in terms of efficiency and quality of the brain and heart anomaly detection in the fetus. The review highlights the importance of integrating multiple data sources, analyzing longitudinal data, and creating larger, more varied datasets for predicting congenital anomalies. The significance of human clinical expertise, interpretability, and prospective validation in real-world clinical settings are also emphasized.