Time-scale image analysis for detection of fetal electrocardiogram
The model presented in this paper is founded on the equivalence between the signal under investigation and its time-scale distribution (TSI). Therefore, separating the signal into multiple sources requires differentiating this TSI into multiple independent TSIs that are intended to represent the ind...
Saved in:
Published in | Multimedia tools and applications Vol. 83; no. 13; pp. 39755 - 39777 |
---|---|
Main Authors | , , |
Format | Journal Article |
Language | English |
Published |
New York
Springer US
01.04.2024
Springer Nature B.V |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The model presented in this paper is founded on the equivalence between the signal under investigation and its time-scale distribution (TSI). Therefore, separating the signal into multiple sources requires differentiating this TSI into multiple independent TSIs that are intended to represent the individual sources. For instance, when applying the continuous wavelet transform to an electrocardiogram signal that simultaneously encompasses the electrical activities of both the fetal hearts (FECG) and the maternal heart (MECG), a highly intricate energy domain is generated, comprising distinct energy sub-domains. These sub-domains can be effectively separated, extracted, and transformed using a segmentation method, resulting in multiple time-scale images that can be analytically inverted into meaningful electrical signals. The presented method will permit the extraction of the fetal contribution to the FECG with optimal filtering of noise and undesirable electrical interference. Furthermore, the provided algorithms yield unequivocal numerical results, as demonstrated through their application to internationally recognized databases, such as the Daisy database and the Fetal ECG Synthetic database from PhysioNet. The obtained Signal-to-Noise Ratio (SNR) and Fetal R Peak Detection Accuracy (FRPDA) values serve to illustrate the efficacy of these techniques in isolating FECG signals from unwanted components and noise, thus enhancing the analysis and interpretation of fetal cardiac health. |
---|---|
ISSN: | 1573-7721 1380-7501 1573-7721 |
DOI: | 10.1007/s11042-023-17165-0 |