An algorithm for the analysis of fetal ECGs from 4-channel non-invasive abdominal recordings

Costanzo Di Maria, Wenfeng Duan, Marjan Bojarnejad, Fan Pan, Susan King, Dingchang Zheng, Alan Murray, Philip Langley

Research output: Contribution to journalConference articlepeer-review

17 Citations (Scopus)


The fetal ECG (fECG) is one of the most valuable tools for monitoring the health of the fetus throughout pregnancy. However, its clinical use has been limited by the difficulty in analysing such non-invasive fECG recordings. The aim of this study was to develop a robust algorithm for the analysis of 4-channel abdominal fECG recordings and test its performance in the Computing in Cardiology Physionet Challenge 2013. Signals were pre-processed by a combination of frequency filtering and wavelet de-noising. Adaptive cancellation of the maternal ECG (mECG) was performed using maternal QRS time markers obtained from the principal component containing the largest mECG. Following further wavelet de-noising of the residuals, the fetal QRS time markers were computed with a local peak detection algorithm from the first principal component. The derived fetal HR (event 4) and fetal RR (event 5) time series were compared to the reference values obtained from a scalp electrode signal. This algorithm scored 223.23 for Challenge event 4 and 19.34 for Challenge event 5, outperforming the sample algorithm.
Original languageEnglish
Pages (from-to)305-308
Number of pages4
JournalComputing in Cardiology
Publication statusPublished - 16 Jan 2014
Externally publishedYes
Event2013 40th Computing in Cardiology Conference, CinC 2013 - Zaragoza, Spain
Duration: 22 Sept 201325 Sept 2013

ASJC Scopus subject areas

  • Computer Science(all)
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'An algorithm for the analysis of fetal ECGs from 4-channel non-invasive abdominal recordings'. Together they form a unique fingerprint.

Cite this