Advanced Source Reconstruction and Volume Conductor Modeling for Fetal Magnetocardiography

Abstract Fetuses that are identified with cardiac hypotrophy, hypertension and metabolic anomalies have higher risk of suffering from various health problems in their later life. Therefore, the early detection of congenital heart anomalies is critical for monitoring or prompt interventions, which can reduce the risks of congestive heart failure. Compared to adult cardiac monitoring, fetal electrophysiological heart monitoring using fetal ECG is extremely difficult due to the low signal amplitude and interferences from the maternal cardiac signal and to the complex environment inside the mother's womb. This problem is even worse in conditions such as diabetic pregnancies because of further signal reduction due to maternal obesity. At the same time, the prevalence of congenital heart anomalies is higher for fetuses of diabetic mothers. The purpose of this thesis is to develop and test fetal magnetocardiography (fMCG) techniques as an alternative diagnostic tool for the detection and monitoring of the fetal heart. fMCG is a novel technique that records the magnetic fields generated by the fetal heart's electric activity. From the aspect of signal processing, magnetic signals generated by the fetal heart are less affected by the low electrical conductivity of the surrounding fetal and maternal tissues compared to the electric signals recorded over the maternal abdomen, and can provide reliable recordings as early as 12 weeks of gestation. However, the fetal heart signals recorded with an array of magnetic sensors at a small distance from the maternal abdomen are affected by the source-to-sensor distance as well as by the geometry of the volume conductor, which is variable in different subjects or in the same subject when recordings are made at different gestational ages. The scope of this thesis is to develop a novel methodology for modeling the fetal heart and volume conductor and to use advanced source reconstruction techniques that can reduce the effect of these confounding factors in evaluating heart magnetic signals. Furthermore, we aim to use these new methods for developing a normative database of fMCG metrics at different gestational ages and test their reliability to detect abnormal patterns of cardiac electrophysiology in pregnancies complicated by maternal diabetes. In the first part of the thesis, we review three current fetal heart monitoring modalities, including fetal electrocardiography (ECG), ultrasonography, and fetal magnetocardiography (fMCG). The advantages and drawbacks of each technique are comparatively discussed. Finally, we discuss the developmental changes of fetal heart through gestation as well as the electromagnetic characteristics of the fetal cardiac activation