Survey on Cardiotocography Feature Extraction Algorithms for Foetal Welfare Assessment

Abstract

Abstract-Since its inception forty years ago as a way to control birth process, the cardiotocograph (CTG) has emerged over time and became the undisputed leader worldwide of noninvasive intrapartum foetal monitoring systems. The CTG signals conveying a lot of information, it is very difficult to interpret them and act accordingly even for specialists; hence, researchers have started looking for characteristics which could be correlated with a particular pathological state of the foetus. Thereby, many features appeared in the literature, ranging from the most common ones to artificially generated features, and computed using a wide variety of signal processing-based analysis tools: time scale, spectral or non-linear analysis, to name but a few. This survey paper, presents in a hierarchical order the most common processing steps of a CTG signal and focuses primarily on the feature extraction methods for foetal heart rate (FHR) analysis reported in the literature during the last decade. Also, some feature classification methods are reported before a brief discussion which concludes this work.

Information Theory to Probe Intrapartum Fetal Heart Rate Dynamics

“…This has triggered a large amounts of researches world wide aiming both to compute in a reproducible and objective way the FIGO criteria [2], and to devise new signal processing inspired features to characterize FHR temporal dynamics (cf. [6,7] for reviews).…”

“…Applied to a large (1404 subjects) and documented database of FHR data, collected in a French academic hospital, it is shown that i) auto mutual information outperforms ApEn and SampEn for acidosis detection in the first stage of labor and continues to yield the best performance in the second stage; ii) Shannon entropy increases as labor progresses, and is always much larger in the second stage; iii) babies suffering from fetal acidosis additionally show more structured temporal dynamics than healthy ones and that this progressive structuration can be used for early acidosis detection.referred to e.g. [6,7] for overviews. There has also been several attempts to combine features different in nature by doing multivariate classification using supervised machine learning strategies (cf.…”

“…referred to e.g. [6,7] for overviews. There has also been several attempts to combine features different in nature by doing multivariate classification using supervised machine learning strategies (cf.…”

Independent Analysis of Decelerations and Resting Periods through CEEMDAN and Spectral-Based Feature Extraction Improves Cardiotocographic Assessment

“…Clinicians visually examine these recordings based on several morphological signal patterns defined in medical guidelines [1][2][3]. Unfortunately, CTG interpretation by this methodology has shown serious disadvantages, such as extensive intra-and inter-observer disagreement, which lead to poor interpretation objectivity and reproducibility [4][5][6].…”

Analysis of the foetal heart rate in cardiotocographic recordings through a progressive characterization of decelerations

“…However, guidelines still lack consensus in many key aspects and are considered too complex and ES do not show concrete evidences on improving foetal welfare assessment [4]. This is why many works have been proposed for extracting hidden patterns from the FHR signal using signal processing techniques, based mainly on FFT, autoregressive modelling, time-varying analysis and wavelets [5]. However, traditional visual CTG analysis is still the primary method to predict foetal welfare in many delivery units.…”