Artificial Intelligence for the Prenatal Ultrasound Diagnosis of Congenital Heart Disease: A Narrative Review

Objective: Congenital heart disease (CHD) is the most prevalent congenital anomaly, imposing a significant burden in morbidity and mortality. Recent advances in artificial intelligence (AI) have introduced numerous new tools to fetal cardiac ultrasound, including automated generation of fetal cardiac planes and identification of specific CHD diagnostic views. Mechanism: Through a narrative review of literature, we described AI technology on automated CHD detection, lesion identification, and associated challenges, such as training datasets and image segmentation. Findings in Brief: The search provided 28 eligible studies. Conclusions: Artificial intelligence seems to be a promising tool to help clinicians in daily clinical activity: it can automate the detection of standard cardiac planes and assist in identifying abnormalities. The main advantages that emerged from this review are related to the reduction of inter- and intra-operator variability, improvement of overall diagnostic performance and accuracy. However, nowadays, its integration into routine clinical practice gives rise to several issues.