Advanced echocardiographic imaging of the fetal heart
Congenital heart disease (CHD) is one of the leading causes of infant morbidity and mortality, accounting for 35-50% of deaths within the first year after birth. Prenatal diagnosis provides a valuable opportunity to intervene and alter the natural history of cardiac diseases. However, accurate prenatal diagnosis of CHD remains suboptimal, with <45% of cases being detected, mostly relying on sonographer training and clinical experience. Over the last decade, 3D fetal echocardiography has been proposed as a valuable tool for fetal heart examination, to overcome the limitations of conventional 2D echo. However, acquisition of consistent, high-quality 3D data remains problematic due to the uncontrollable nature of fetal orientation and motion. Cardiovascular magnetic resonance (CMR) has also recently demonstrated significant potential as a diagnostic tool, but, despite being superior in quality and structural detail, fetal CMR is still under development and remains less accessible than echocardiography. Finally, although not diagnostic, micro-CT has become a valuable tool to understand and study the structure of the fetal heart post-mortem. The plan for this project is to radically transform fetal imaging to enable accurate, robust and reproducible diagnosis of CHD through a fundamentally different approach to prenatal diagnosis: rather than addressing the limitations of each individual image modality, we plan to leverage their strongest points through the use of mathematical and computational models: 3D whole-heart models, obtained from the combination of previously acquired data (CMR and micro-CT), will be used to guide 2D/3DE acquisitions and to enrich the diagnosis information of the acquired images.
Aims and Objectives:
With the hypothesis that the complementary use of multiple imaging modalities will lead to improved prenatal diagnosis and a better understanding of CHD, the global aim of this project is to improve the clinical prenatal diagnosis of CHD through the development of novel automated computational imaging tools leading to advanced echocardiographic images. This will be achieved by coupling an on-going 2D/3DE acquisition with offline-built, spatially consistent, gestational age-specific 3D models of the fetal heart obtained from multi-modal images.