Machine Learning in Fetal Cardiology: What to Expect

Abstract

In fetal cardiology, imaging (especially echocardiography) has demonstrated to help in the diagnosis and monitoring of fetuses with a compromised cardiovascular system potentially associated with several fetal conditions. Different ultrasound approaches are currently used to evaluate fetal cardiac structure and function, including conventional 2-D imaging and M-mode and tissue Doppler imaging among others. However, assessment of the fetal heart is still challenging mainly due to involuntary movements of the fetus, the small size of the heart, and the lack of expertise in fetal echocardiography of some sonographers. Therefore, the use of new technologies to improve the primary acquired images, to help extract measurements, or to aid in the diagnosis of cardiac abnormalities is of great importance for optimal assessment of the fetal heart. Machine leaning (ML) is a computer science discipline focused on teaching a computer to perform tasks with specific goals without explicitly programming the rules on how to perform this task. In this re-view we provide a brief overview on the potential of ML techniques to improve the evaluation of fetal cardiac function by optimizing image acquisition and quantification/segmentation, as well as aid in improving the prenatal diagnoses of fetal cardiac remodeling and abnormalities.

The left outflow tract in foetal echocardiography screening: an audit of practices and quality criteria.

“…Analysis of the foetal heart remains challenging, mainly due to its complex anatomy, the small size of the areas studied, heart movements, involuntary foetal movements, and the lack of expertise in foetal echocardiography among some sonographers. Therefore, the use of new Artificial Intelligence-supported technologies aimed at making screening echocardiography practices safer seems likely to be the next step towards precision medicine in the future (25) (31). Ethics Approval: As this was non-interventional retrospective and prospective research involving nonhuman data from a study in the health field, a simplified MR004 procedure was granted with an agreement to comply with the National Commission for Information Technology and Civil Liberties (CNIL) (CNIL Reference 2219789 v 0) and registration of the project on the Register of studies without CNIL authorization.…”

“…It is with this in mind that Artificial Intelligence (AI) has been developed. Currently, the most relevant model in automated learning of image analysis is the convolutional neural network (25) (26). AI could thus help the practitioner by carrying out a real-time audit of the images provided (presence or absence and quality), helping to obtain recommended quality anatomical images (27), but also in terms of improving image resolution (28)(29) (30).…”

Right or wrong? Looking through the retrospectoscope to analyse predictions made a decade ago in prenatal diagnosis and fetal surgery

“…24 Artificial intelligence in the near future may overcome this inherent limitation of prenatal ultrasonography. 25,26 Furthermore the detection of fetal anomalies may move even earlier with advances in transvaginal imaging and operator experience 27 especially as novel imaging technologies are facilitating excellent post-mortem examination even in very small fetuses. 28 One might have expected that the place of fetal ultrasound in prenatal diagnosis will be diminished with the rapid advances being made with molecular genetic prenatal diagnosis.…”

Local Preserving Class Separation Framework to Identify Gestational Diabetes Mellitus Mother Using Ultrasound Fetal Cardiac Image

“…The CAD in this study is based on ultrasound (US) imaging. Many recent studies have proposed accurate CAD models using US images [15]- [23]. Physicians can typically distinguish or identify the disease by manual interpretation of the images, but the process is intensive, induces fatigability, and is prone to bias and human error.…”