Effects of fetal and maternal breathing on the ultrasonic Doppler signal due to fetal heart movement

KUBAI: Sensor Fusion for Non-Invasive Fetal Heart Rate Tracking

OBJECTIVE

Fetal heart rate (FHR) is critical for perinatal fetal monitoring. However, motions, contractions and other dynamics may substantially degrade the quality of acquired signals, hindering robust tracking of FHR. We aim to demonstrate how use of multiple sensors can help overcome these challenges.

METHODS

We develop KUBAI1, a novel stochastic sensor fusion algorithm, to improve FHR monitoring accuracy. To demonstrate the efficacy of our approach, we evaluate it on data collected from gold standard large pregnant animal models, using a novel non-invasive fetal pulse oximeter.

RESULTS

The accuracy of the proposed method is evaluated against invasive ground-truth measurements. We obtained below 6 beats-per-minute (BPM) root-mean-square error (RMSE) with KUBAI, on five different datasets. KUBAI's performance is also compared against a single-sensor version of the algorithm to demonstrate the robustness due to sensor fusion. KUBAI's multi-sensor estimates are found to give overall 23.5% to 84% lower RMSE than single-sensor FHR estimates. The mean ± SD of improvement in RMSE is 11.95 ±9.62 BPM across five experiments. Furthermore, KUBAI is shown to have 84% lower RMSE and  ∼ 3 times higher R2 correlation with reference compared to another multi-sensor FHR tracking method found in literature.

CONCLUSION

The results support the effectiveness of KUBAI, the proposed sensor fusion algorithm, to non-invasively and accurately estimate fetal heart rate with varying levels of noise in the measurements.

SIGNIFICANCE

The presented method can benefit other multi-sensor measurement setups, which may be challenged by low measurement frequency, low signal-to-noise ratio, or intermittent loss of measured signal.

Robust Fetal Heart Rate Tracking through Fetal Electrocardiography (ECG) and Photoplethysmography (PPG) Fusion. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2023;2023:1-4. [PMID: 38083436 DOI: 10.1109/embc40787.2023.10341068] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Fetal electrocardiogram (fECG) or photoplethysmogram (fPPG) devices are being developed for fetal heart rate (FHR) monitoring. However, deep tissue sensing is challenged by low fetal signal-to-noise ratio (SNR). Data quality is easily degraded by motion, or interference from maternal tissues and data losses can happen due to communication faults. In this paper, we propose to combine fECG and fPPG measurements in order to increase robustness against such dynamic challenges and increase FHR estimation accuracy. To the author's knowledge the fusion of two sensory data types (fECG, fPPG) has not been investigated for FHR tracking purposes in the literature. The proposed methods are evaluated on real-world data captured from gold-standard large pregnant animal experiments. A particle filtering algorithm with sensor fusion in the measurement likelihood, called KUBAI, is used to estimate FHR. Fusion of PPG&ECG data resulted in 36.6% improvement in root-mean-square-error (RMSE) and 20.3% improvement in R2 correlation between estimated and reference FHR values compared to single sensor-type (PPG-only or ECG-only) data. We demonstrate that using different types of sensory data improves the robustness and accuracy of FHR tracking.

Multi-Detector Heart Rate Extraction Method for Transabdominal Fetal Pulse Oximetry. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2021;2021:1072-1075. [PMID: 34891473 PMCID: PMC10631454 DOI: 10.1109/embc46164.2021.9630946]

Intrapartum fetal well-being assessment relies on fetal heart rate (FHR) monitoring. Studies have shown that FHR monitoring has a high false-positive rate for detecting fetal hypoxia during labor and delivery. A transabdominal fetal pulse oximeter device that measures fetal oxygen saturation non-invasively through NIR light source and photodetectors could increase the accuracy of hypoxia detection. As light travels through both maternal and fetal tissue, photodetectors on the surface of mother's abdomen capture mixed signals comprising fetal and maternal information. The fetal information should be extracted first to enable fetal oxygen saturation calculation. A multi-detector fetal signal extraction method is presented in this paper where adaptive noise cancellation is applied to four mixed signals captured by four separate photodetectors placed at varying distances from the light source. As a result of adaptive noise cancellation, we obtain four separate FHR by peak detection. Weighting, outlier rejection and averaging are applied to these four fetal heart rates and a mean FHR is reported. The method is evaluated in utero on data collected from hypoxic lamb model. Ground truth for FHR is measured through hemodynamics. The results showed that using multi-detector fetal signal extraction gave up to 18.56% lower root-mean-square FHR error, and up to 57.87% lower maximum absolute FHR error compared to single-detector fetal signal extraction.

Automatic fetal movement recognition from multi-channel accelerometry data

BACKGROUND AND OBJECTIVE

Significant health care resources are allocated to monitoring high risk pregnancies to minimize growth compromise, reduce morbidity and prevent stillbirth. Fetal movement has been recognized as an important indicator of fetal health. Studies have shown that 25% of pregnancies with decreased fetal movement in the third trimester led to poor outcomes at birth. The studies have also shown that maternal perception of fetal movement is highly subjective and varies from person to person. A non-invasive system for fetal movement detection that can be used outside hospital would represent an advance in at-home monitoring of at-risk pregnancies. This is a challenging task that requires the use of advanced signal processing techniques to differentiate genuine fetal movements from contaminating artefacts.

METHODS

This manuscript proposes a novel algorithm for automatic fetal movement recognition using data collected from wearable tri-axial accelerometers strategically placed on the maternal abdomen. The novelty of the work resides in the efficient removal of artefacts and in distinctive feature extraction. The proposed algorithm used independent component analysis (ICA) for dimensionality reduction and artefact removal. A supplemental technique based on discrete wavelet transform (DWT) was also used to remove artefacts.

RESULTS

To identify fetal movements, 31 features were extracted from the acceleration data. Based on these features, several classifiers were used to distinguish fetal from non-fetal movements. Robustness of the classifiers was tested for various concentrations of artefacts in the classification data. The best performance was achieved by Bagging classifier algorithm, with random forest as its basis classifier, yielding an accuracy ranging from 87.6% to 95.8% depending on the artefact concentration level.

CONCLUSIONS

A high performance detection of fetal movements can be achieved using accelerometery-based systems suitable for long-term monitoring.

Multidimensional Ultrasound Doppler Signal Analysis for Fetal Activity Monitoring

Fetal activity parameters such as movements, heart rate and the related parameters are essential indicators of fetal wellbeing, and no device provides simultaneous access to and sufficient estimation of all of these parameters to evaluate fetal health. This work was aimed at collecting these parameters to automatically separate healthy from compromised fetuses. To achieve this goal, we first developed a multi-sensor-multi-gate Doppler system. Then we recorded multidimensional Doppler signals and estimated the fetal activity parameters via dedicated signal processing techniques. Finally, we combined these parameters into four sets of parameters (or four hyper-parameters) to determine the set of parameters that is able to separate healthy from other fetuses. To validate our system, a data set consisting of two groups of fetal signals (normal and compromised) was established and provided by physicians. From the estimated parameters, an instantaneous Manning-like score, referred to as the ultrasonic score, was calculated and was used together with movements, heart rate and the associated parameters in a classification process employing the support vector machine method. We investigated the influence of the sets of parameters and evaluated the performance of the support vector machine using the computation of sensibility, specificity, percentage of support vectors and total classification error. The sensitivity of the four sets ranged from 79% to 100%. Specificity was 100% for all sets. The total classification error ranged from 0% to 20%. The percentage of support vectors ranged from 33% to 49%. Overall, the best results were obtained with the set of parameters consisting of fetal movement, short-term variability, long-term variability, deceleration and ultrasound score. The sensitivity, specificity, percentage of support vectors and total classification error of this set were respectively 100%, 100%, 35% and 0%. This indicated our ability to separate the data into two sets (normal fetuses and pathologic fetuses), and the results highlight the excellent match with the clinical classification performed by the physicians. This work indicates the feasibility of detecting compromised fetuses and also represents an interesting method of close fetal monitoring during the entire pregnancy.

Fetal heart rate extraction from abdominal electrocardiograms through multivariate empirical mode decomposition

Assessment of fetal heart rate (FHR) and fetal heart rate variability (fHRV) reveals important information about fetal well-being, specifically in high risk pregnancies. Abdominal electrocardiogram (abdECG) recording is a non-invasive method to capture fetal electrocardiograms. In this paper, we propose a methodology to extract FHR (fetal RR time series) from the abdECG recordings using the recently introduced multivariate empirical mode decomposition (MEMD) technique. MEMD breaks a signal into a finite set of intrinsic mode functions (IMFs). First, elimination of the noisier abdECG channels, based on comparison of similar indexed IMFs that were obtained through the MEMD technique, is conducted. Thereafter, denoising of the remaining abdECG channels is performed by eliminating certain similar indexed IMFs. The unwanted mother QRS complexes are removed from these noise-free abdECG channels, and the candidate fetal R-peaks are detected through a wavelet based approach. The proposed methodology is validated using an open source real-life clinical database. The proposed technique resulted in a high value (0.983) of cross correlation between the detected and true FHR signals.

Empirical Mode Decomposition of simulated and real ultrasonic Doppler signals of periodic fetal activity

Simulated signals comprising components (trains of Gaussian packets) resulting from cardiac movements and from pseudorespiratory movements with added white noise were submitted to Empirical Mode Decomposition. The increase of sampling frequency fs (from 0.5 kHz to 5 kHz) for given signal to noise ratio SNR moves signal components toward higher order intrinsic mode functions (IMFs) and increases their number. The increase of the SNR (from -5 dB to 10 dB, fixed fs) moves the signal components to lower order IMFs. The separation of components is most efficient for SNR≥5 dB and fs not exceeding 1 kHz, for lower SNRs fs should be at least 2 kHz. SNR=∞ results in erroneous decomposition and therefore limited noise level is beneficial. Recommended number of sifting iterations is 10. Fetal data obtained using 2 MHz emission frequency and sampled at 2 kHz were decomposed. The cardiac signal always appears in IMF3, frequently also in IMF1 and IMF2. The pseudobreathing signal, appearing mainly in IMF4-6, is easy to separate. Signals resulting from fetal displacements due to maternal respiration appear in IMF7 or IMF8. The EMD performs better than the classic linear filtering as a tool for separation of the pseudorespiration signals and provides inferior results in terms of separation of the cardiac signals.

A PC-aided optical foetal heart rate detection system

Safe monitoring of foetal heart rate is a valuable tool for the healthy evolution and wellbeing of both foetus and mother. This paper presents a non-invasive optical technique that allows for foetal heart rate detection using a photovoltaic infrared (IR) detector placed on the mother's abdomen. The system presented here consists of a photoplethysmography (PPG) circuit, abdomen circuit and a personal computer equipped with MATLAB. A near IR beam having a wavelength of 880 nm is transmitted through the mother's abdomen and foetal tissue. The received abdominal signal that conveys information pertaining to the mother and foetal heart rate is sensed by a low noise photodetector. The PC receives the signal through the National Instrumentation Data Acquisition Card (NIDAQ). After synchronous detection of the abdominal and finger PPG signals, the designed MATLAB-based software saves, analyses and extracts information related to the foetal heart rate. Extraction is carried out using recursive least squares adaptive filtration. Measurements on eight pregnant women with gestational periods ranging from 35-39 weeks were performed using the proposed system and CTG. Results show a correlation coefficient of 0.978 and a correlation confidence interval between 88-99.6%. The t test results in a p value of 0.034, which is less than 0.05. Low power, low cost, high signal-to-noise ratio, reduction of ambient light effect and ease of use are the main characteristics of the proposed system.

Applicability of adaptive noise cancellation to fetal heart rate detection using photoplethysmography

In this paper, an approach based on adaptive noise cancellation (ANC) is evaluated for extraction of the fetal heart rate using photoplethysmographic signals from the maternal abdomen. A simple optical model is proposed in which the maternal and fetal blood pulsations result in emulated signals where the lower SNR limit (fetal to maternal) is -25dB. It is shown that a recursive least-squares algorithm is capable of extracting the peaks of the fetal PPG from these signals, for typical values of maternal and fetal tissues.

Automatic detection of ultrasonic Doppler signal episodes resulting from fetal breathing movements

A method for automatic detection of fetal breathing movements has been proposed, based on the time-frequency structure of the corresponding continuous wave ultrasonic Doppler signals. The method uses spectral analysis of the envelope of the directional Doppler signal and cross-correlation analysis of both directional envelopes. Detection rule comprises the following criteria: presence of the peak in the envelope spectrum and of the adequate signal level in the frequency range corresponding to the fetal breathing rhythm, the peak value and the position limits of the peak of the cross-correlation coefficient of the both directional envelopes. The effect of the criteria setting on the rule performance and the tradeoff between the specificity and sensitivity was investigated. The rule is most sensitive to the threshold value of the cross-correlation coefficient of the envelopes. The limits of the position of this peak are crucial for the distinction between the breathing episodes and hiccups. The optimal settings of the criteria, resulting in average sensitivity and specificity exceeding, respectively, 0.70 and 0.80, are proposed.

The Actifetus system: a multidoppler sensor system for monitoring fetal movements

Fetal heart rate (FHR) monitoring is a crucial part of monitoring at-risk pregnancies and labor. Its aim is to detect any abnormalities that might indicate acute fetal distress and a need for rapid treatment to avoid death or serious sequelae, including cerebral handicap. The use of fetal biophysical profiles in high-risk pregnancies (gravidic hypertension, in utero infection, etc.) helps to distinguish healthy fetuses from those with chronic conditions. Fetal biophysical profile scores have been developed that integrate five biophysical parameters, one of which is derived from the FHR. The major parameters detected are the rate of fetal movements, fetal tone, fetal breathing movement and amniotic fluid volume. All of those parameters except FHR are obtained by prolonged echographic observation and cannot be used routinely. We developed in this study a new multigate multitransducer pulsed Doppler system for survey of fetal behavior. Fast Fourier transform and autocorrelation function have been used for processing and analyzing ultrasonic Doppler signals generated by fetal movements. Several parameters are analyzed in each of the 12 x 5 = 60 Doppler gates: amplitude of signals reflected by moving fetal structures, velocity, direction and amplitude of displacement of fetal structure (heart, chest, limbs). From these parameters it is possible to calculate FHR and characterize fetal activity. Preliminary in vivo results obtained in 15 pregnant women (30 to 36 wk) are very encouraging but they have yet to be confirmed in future studies. These results also demonstrate the advantages of transducers designed for improved fetal movement detection. The algorithms needs to be precise enough to allow the Actifetus system to function in real time. We now have at our disposal some algorithms that succeed in quantifying FHR and fetal movements with a signal from a given sensor at a given depth. This study confirms the feasibility of monitoring fetal movements by the Actifetus system and demonstrates the importance of the characterization of fetal rhythms (and fetal behavior). The Actifetus system will serve as a new mean for studying fetal response to environment and detecting anomalies related to fetal suffering.

A new algorithm for detecting fetal heart rate using ultrasound Doppler signals

Although the fetal heart rate monitoring using ultrasound is widely used it is still not optimized for automatic measurements due to the complexity of the Doppler signal. This paper presents a new fetal heart rate (FHR) detecting algorithm, using sampling auto-correlation approach. The results obtained using the custom-built ultrasonic Doppler fetal heart rate monitoring system are presented and confirm the validity of the method.

The DopFet system: a new ultrasonic Doppler system for monitoring and characterization of fetal movement

To enable the investigation of fetal movement in a manner similar to fetal heart rate (FHR) monitoring we have developed an apparatus (the DopFet system) that consists of a pair of miniature sensors, a 2-MHz continuous-wave directional Doppler electronic module and a laptop personal computer. One of the sensors is aimed at the fetal limbs and the other at the thorax to detect heart and upper body movements. The signals are analyzed, presented in real-time and postprocessed by software developed by us. The postprocessing software computes a number of parameters (the DopFet parameters) describing fetal movement. These parameters can be divided into two categories: parameters that describe the quantity of fetal movement (i.e., number of movements) and parameters that describe qualitative aspects of fetal movement (i.e., average movement duration). Future studies using the DopFet system will be aimed at discovering which of these parameters or combination of parameters is the best indicator of fetal well-being. We present an example of a 0.5 h recording and the results of testing on 23 volunteer mothers. These results show good sensitivity of the system compared to real-time ultrasound (US). The system detects 96% of rolling movements, 100% of flexion movements and 97% of leg movements.