Automated Detection of Congenital Heart Disease in Fetal Ultrasound Screening
This addresses the challenge of limited detection rates in prenatal screening due to reliance on human expertise and high case volumes, offering an incremental improvement for medical imaging in healthcare.
The paper tackles the problem of detecting congenital heart disease in fetal ultrasound screening by proposing a deep learning pipeline that uses an auxiliary view classification task to bias features toward cardiac structures, resulting in improved F1-scores from 0.72 and 0.77 to 0.87 and 0.85 for healthy and CHD classes.
Prenatal screening with ultrasound can lower neonatal mortality significantly for selected cardiac abnormalities. However, the need for human expertise, coupled with the high volume of screening cases, limits the practically achievable detection rates. In this paper we discuss the potential for deep learning techniques to aid in the detection of congenital heart disease (CHD) in fetal ultrasound. We propose a pipeline for automated data curation and classification. During both training and inference, we exploit an auxiliary view classification task to bias features toward relevant cardiac structures. This bias helps to improve in F1-scores from 0.72 and 0.77 to 0.87 and 0.85 for healthy and CHD classes respectively.