Fetal Ultrasound Video

Fetal ultrasound video analysis is rapidly advancing, aiming to automate tasks currently performed by sonographers, such as biometric measurements and gestational age estimation. Research focuses on developing deep learning models, including convolutional neural networks and transformer-based architectures, to analyze spatio-temporal features within ultrasound videos, often incorporating techniques like keyframe selection and plane pose estimation for improved accuracy and efficiency. These advancements promise to improve the speed and consistency of prenatal diagnosis, potentially reducing workload for clinicians and enhancing the quality of care, particularly in resource-constrained settings. Furthermore, research is actively addressing challenges like out-of-distribution sample detection and automated quality assessment to increase the reliability and robustness of these automated systems.