Learning normal appearance for fetal anomaly screening: Application to the unsupervised detection of Hypoplastic Left Heart Syndrome
This work addresses the problem of early and accurate detection of congenital heart disease in newborns, which affects 6-11 per 1000 births, by providing an automated screening tool for medical professionals, though it is incremental as it builds on existing anomaly detection frameworks.
The paper tackled automated detection of cardiac anomalies in fetal ultrasound screening, specifically Hypoplastic Left Heart Syndrome, using an unsupervised approach that learns from normal control patients, achieving an average AUC of 0.81 and improved robustness compared to state-of-the-art methods.
Congenital heart disease is considered as one the most common groups of congenital malformations which affects $6-11$ per $1000$ newborns. In this work, an automated framework for detection of cardiac anomalies during ultrasound screening is proposed and evaluated on the example of Hypoplastic Left Heart Syndrome (HLHS), a sub-category of congenital heart disease. We propose an unsupervised approach that learns healthy anatomy exclusively from clinically confirmed normal control patients. We evaluate a number of known anomaly detection frameworks together with a model architecture based on the $α$-GAN network and find evidence that the proposed model performs significantly better than the state-of-the-art in image-based anomaly detection, yielding average $0.81$ AUC \emph{and} a better robustness towards initialisation compared to previous works.