Baumeister Elena

Elena Baumeister

Bachelor's Thesis

Preprocessing and Standardization Techniques for Comparative Analysis of Fetal ECGs: A step towards Detecting Congenital Heart Defects


Katharina Jäger (M. Sc.), Michael Nissen (M. Sc.), Dr. – Ing. Heike Leutheuser, Prof. Dr. Björn Eskofier


03 / 2024 – 08 / 2024


Congenital heart defects are the most common single-organ malformations. Thereby, about 1.1% of newborns are affected by congenital heart defects (CHD) in Germany [1]. Currently, CHD is prenatally screened by ultrasound and postnatally by pulse oximetry. Due to pulse oximetry, the detection rate is 82% within the first three months after birth, and 12% of CHD are detected by ultrasound prenatally [1]. However, the quality depends on the skills of the sonographer, the fetal position, and the maternal body mass index [2]. Early detection of possible CHD would enable prenatal therapy or required preparation for labor and postnatal interventions [2].

An alternative for prenatal examination could be a non-invasive fetal electrocardiogram (NIFECG). There is evidence in research that it is feasible to detect CHD in NI-FECG [2, 3]. However, due to the lack of research on the normal morphology of a healthy fetal ECG, detecting abnormal fetal ECG morphologies is still a challenging task. For that reason, preprocessing and standardization techniques are needed to normalize the measured electrocardiograms with respect to the fetal position to compare morphologies.

The thesis, therefore, aims to implement an algorithm for standardizing non-invasive fetal ECG signals regarding the fetal position to make waveform analysis and evaluation more comparable. In the next step, the influence of this standardization on fetal ECG morphology characteristics is analyzed. Within this project, the ECG and pulse-wave Doppler signals of the open-access dataset NInFEA: Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research by Pani et al. [4, 5] must be preprocessed, filtered, and evaluated according to existing standardization methods by Lempersz et al. [2,3]. Furthermore, the impact of the standardization on morphology characteristics on the ECG will be considered.


[1] Lindinger, G. Schwedler, and H.-W. Hense, „Prevalence of congenital heart defects in newborns in Germany: Results of the first registration year of the PAN Study,” Klin Padiatr, vol. 222, no. 5, pp. 321-326, (2010).

[2] Lempersz C., van Laar J.O., Clur S.-A.B., Verdurmen K.M., Warmerdam G.J., van der Post J., et al.(2020). “The standardized 12-lead fetal electrocardiogram of the healthy fetus in mid-pregnancy: A cross-sectional study.” PLoS ONE15 (4): e0232606.

[3] Lempersz C., Noben L.,Clur S.-A.B., van den Heuvel E., Zhan Z., Haak M., et al.(2021). „The electrical heart axis of the fetus between 18 and 24 weeks of gestation: A cohort study. PLoS ONE 16(12): e0256115

[4] Sulas, E., Urru, M., Tumbarello, R. et al. „A non-invasive multimodal foetal ECG–Doppler dataset for antenatal cardiology research.” Sci Data 8, 30 (2021).

[5] Pani, D., Sulas, E., Urru, M., Sameni, R., Raffo, L., & Tumbarello, R. (2020). NInFEA: Non-Invasive Multimodal Foetal ECGDoppler Dataset for Antenatal Cardiology Research (version 1.0.0). PhysioNet. (