Auscultation and observation of foetal heart rate (FHR), through various techniques, has always been of crucial importance to ascertain life and well-being of the foetus. Gynaecologists evaluate several characteristics of FHR signal, but foetal heart rate variability (FHRV) is probably the most important feature of the FHR recording, due to its relationship to the autonomic and central nervous systems, as demonstrated by pharmacological experiments in both foetal sheep and adult humans [1]. However, there is a high inter and intra-observer variability [2] in the interpretation of FHR signals and there are no clear standards for the definition and evaluation of FHR changes. This is particularly true when assessing beat-to-beat variability of the heart rate [1, 3, 4]. In frequency domain, as well as adults, most of literature [1, 5, 6] agrees that three bands can be detected in the FHR power spectrum, a very low frequency band (VLF), a low frequency band (LF) and a high frequency band (HF). However, despite to the spread use of FHRV spectral analysis [7, 8, 9, 10], there is not agreement in definition of its components. Some researcher groups, for example, identify the VLF with the range [0 ??? 0.03] Hz [6], while other authors consider as VLF the range [0 ??? 0.04] Hz [11] or [0 ??? 0.05]Hz [1]. Moreover, some authors introduce also a medium frequency band (MF), corresponding to maternal respiration [12]. Besides, differently from what happens in the adult, there are modifications of the baseline, accelerations and decelerations, which make more difficult or even impossible FHRV definition and identification. Some guidelines in fact state that variability refers to fluctuations in the baseline free from accelerations and decelerations [13]. This is an important limit in clinical routine since variability in correspondence of these FHR alterations has always been regarded as particularly significant in terms of prognostic value [14, 15]. Aim of this work was to propose the computation of FHRV signal as difference between FHR and floatingline, which is an imaginary median line of FHR signal [16, 17] and that hence includes accelerations and decelerations. Further, we characterised floatingline in frequency domain, and, to this purpose, a characterisation of accelerations and decelerations has been necessary, being the literature very poor at this regard. Finally, a procedure to estimate floatingline in time domain was proposed.
Floatingline estimation in FHR signal analysis
Romano M.;
2011-01-01
Abstract
Auscultation and observation of foetal heart rate (FHR), through various techniques, has always been of crucial importance to ascertain life and well-being of the foetus. Gynaecologists evaluate several characteristics of FHR signal, but foetal heart rate variability (FHRV) is probably the most important feature of the FHR recording, due to its relationship to the autonomic and central nervous systems, as demonstrated by pharmacological experiments in both foetal sheep and adult humans [1]. However, there is a high inter and intra-observer variability [2] in the interpretation of FHR signals and there are no clear standards for the definition and evaluation of FHR changes. This is particularly true when assessing beat-to-beat variability of the heart rate [1, 3, 4]. In frequency domain, as well as adults, most of literature [1, 5, 6] agrees that three bands can be detected in the FHR power spectrum, a very low frequency band (VLF), a low frequency band (LF) and a high frequency band (HF). However, despite to the spread use of FHRV spectral analysis [7, 8, 9, 10], there is not agreement in definition of its components. Some researcher groups, for example, identify the VLF with the range [0 ??? 0.03] Hz [6], while other authors consider as VLF the range [0 ??? 0.04] Hz [11] or [0 ??? 0.05]Hz [1]. Moreover, some authors introduce also a medium frequency band (MF), corresponding to maternal respiration [12]. Besides, differently from what happens in the adult, there are modifications of the baseline, accelerations and decelerations, which make more difficult or even impossible FHRV definition and identification. Some guidelines in fact state that variability refers to fluctuations in the baseline free from accelerations and decelerations [13]. This is an important limit in clinical routine since variability in correspondence of these FHR alterations has always been regarded as particularly significant in terms of prognostic value [14, 15]. Aim of this work was to propose the computation of FHRV signal as difference between FHR and floatingline, which is an imaginary median line of FHR signal [16, 17] and that hence includes accelerations and decelerations. Further, we characterised floatingline in frequency domain, and, to this purpose, a characterisation of accelerations and decelerations has been necessary, being the literature very poor at this regard. Finally, a procedure to estimate floatingline in time domain was proposed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
