Non-invasive monitoring of fetal cardiac activity is of great clinical interest to assess fetal health. To date, however, difficulties in detecting fetal beats from abdominal mother recordings prevented the possibility of obtaining reliable results. In this study a multi-step approach for the analysis of non-invasive fetal ECG is proposed. The first steps concern the pre-processing stages of baseline removal and power line interference canceling. The successive operations are: Independent Component Analysis (ICA) for maternal ECG extraction; mother QRS detection; maternal ECG canceling using a PQRST approximation obtained by weighted Singular Value Decomposition (SVD); second ICA applied to enhance the fetal ECG signal; fetal QRS detection. The results obtained in Physionet Challenge 2013 on the test sets are expressed as two scores (HRmse and RRrmse) measuring respectively the matching between the reference annotations of fetal HR and RR time series and those estimated with the developed software. The results obtained on the learning set are: sensitivity=99.4%, positive predictive accuracy=99.2% and HRmse=1.52 bpm2, RRrmse=2.11 ms. The scores for the open test set are: HRmse=34.0 bpm2, RRrmse=5.10 ms. The scores for the hidden test (open source section) are: HRmse=187 bpm2, RRrmse=21.0 ms.

A multi-step approach for non-invasive fetal ECG analysis

VARANINI, MAURIZIO;TARTARISCO, GENNARO;MACERATA, ALBERTO MARCO MARIA;PIOGGIA, GIOVANNI;
2013

Abstract

Non-invasive monitoring of fetal cardiac activity is of great clinical interest to assess fetal health. To date, however, difficulties in detecting fetal beats from abdominal mother recordings prevented the possibility of obtaining reliable results. In this study a multi-step approach for the analysis of non-invasive fetal ECG is proposed. The first steps concern the pre-processing stages of baseline removal and power line interference canceling. The successive operations are: Independent Component Analysis (ICA) for maternal ECG extraction; mother QRS detection; maternal ECG canceling using a PQRST approximation obtained by weighted Singular Value Decomposition (SVD); second ICA applied to enhance the fetal ECG signal; fetal QRS detection. The results obtained in Physionet Challenge 2013 on the test sets are expressed as two scores (HRmse and RRrmse) measuring respectively the matching between the reference annotations of fetal HR and RR time series and those estimated with the developed software. The results obtained on the learning set are: sensitivity=99.4%, positive predictive accuracy=99.2% and HRmse=1.52 bpm2, RRrmse=2.11 ms. The scores for the open test set are: HRmse=34.0 bpm2, RRrmse=5.10 ms. The scores for the hidden test (open source section) are: HRmse=187 bpm2, RRrmse=21.0 ms.
9781479908844
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11568/627871
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact