Learning normal appearance for fetal anomaly screening: Application to the unsupervised detection of Hypoplastic Left Heart Syndrome

Elisa ChotzoglouImperial College London, UK, Thomas DayKing’s College London, UK, Jeremy TanImperial College London, UK, Jacqueline MatthewKing’s College London, UK, David LloydKing’s College London, UK, Reza RazaviKing’s College London, UK, John SimpsonKing’s College London, UK, Bernhard KainzImperial College London, UK
September 2021 issue
Publication date: 2021/10/22
PDF · arXiv

Abstract

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 new 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 and a better robustness towards initialisation compared to previous works.

Keywords

fetal screening · detection · unsupervised learning

Bibtex @article{melba:2021:012:chotzoglou, title = "Learning normal appearance for fetal anomaly screening: Application to the unsupervised detection of Hypoplastic Left Heart Syndrome", authors = "Chotzoglou, Elisa and Day, Thomas and Tan, Jeremy and Matthew, Jacqueline and Lloyd, David and Razavi, Reza and Simpson, John and Kainz, Bernhard", journal = "Machine Learning for Biomedical Imaging", volume = "1", issue = "September 2021 issue", year = "2021" }

2021:012 cover