Publicly Available Fetal Ultrasound Databases: Current Status and Future Directions

Fetal ultrasound imaging plays a vital role in antenatal care, yet the standardization and accessibility of public databases remain critical challenges in advancing research and clinical applications. This review analyzes 14 publicly available fetal ultrasound databases (2017-2024) identified through a comprehensive search of Web of Science, Scopus, and Google Scholar. Our analysis reveals a diverse landscape of datasets comprising both static and dynamic imaging data, with sample sizes ranging from 25 video sequences to 15,728 images. While most databases (11/14) focus on 2D imaging, emerging specialized collections include cardiac videos, Doppler traces, and 3D reconstructions. Technical validation demonstrates high accuracy (82-98%) across applications, though significant gaps exist in standardization, pathological case representation, and demographic diversity. The review identifies key trends in database development, including the emergence of phantom-based datasets, increased focus on low-resource settings through transfer learning, and integration of synthetic data generation. We propose strategic recommendations for future database development, emphasizing clinical workflow integration, standardized quality metrics, and inclusive data collection strategies to bridge the gap between technical advancement and practical clinical implementation.