Cardiotocography (CTG) is a non-invasive and important tool for diagnosing fetal distress during pregnancy. To meet the needs of intelligent fetal heart monitoring based on deep learning, this paper proposes a TWD-MOAL deep active learning algorithm based on the three-way decision (TWD) theory and multi-objective optimization Active Learning (MOAL). During the training process of a convolutional neural network (CNN) classification model, the algorithm incorporates the TWD theory to select high-confidence samples as pseudo-labeled samples in a fine-grained batch processing mode, meanwhile low-confidence samples annotated by obstetrics experts were also considered. The TWD-MOAL algorithm proposed in this paper was validated on a dataset of 16 355 prenatal CTG records collected by our group. Experimental results showed that the algorithm proposed in this paper achieved an accuracy of 80.63% using only 40% of the labeled samples, and in terms of various indicators, it performed better than the existing active learning algorithms under other frameworks. The study has shown that the intelligent fetal heart monitoring model based on TWD-MOAL proposed in this paper is reasonable and feasible. The algorithm significantly reduces the time and cost of labeling by obstetric experts and effectively solves the problem of data imbalance in CTG signal data in clinic, which is of great significance for assisting obstetrician in interpretations CTG signals and realizing intelligence fetal monitoring.
Humans ; Pregnancy ; Female ; Cardiotocography/methods* ; Deep Learning ; Neural Networks, Computer ; Algorithms ; Fetal Monitoring/methods* ; Heart Rate, Fetal ; Fetal Distress/diagnosis* ; Fetal Heart/physiology*

Background: Valproic acid (VPA) exposure during pregnancy has been proven to contribute to congenital heart disease (CHD). Our previous findings implied that disruption of planar cell polarity (PCP) signaling pathway in cardiomyocytes might be a factor for the cardiac teratogenesis of VPA. In addition, the teratogenic ability of VPA is positively correlated to its histone deacetylase (HDAC) inhibition activity. This study aimed to investigate the effect of the VPA on cardiac morphogenesis, HDAC1/2/3, and PCP key genes (Vangl2/Scrib/Rac1), subsequently screening out the specific HDACs regulating PCP pathway. Methods: VPA was administered to pregnant C57BL mice at 700 mg/kg intraperitoneally on embryonic day 10.5. Dams were sacrificed on E15.5, and death/absorption rates of embryos were evaluated. Embryonic hearts were observed by hematoxylin-eosin staining to identify cardiac abnormalities. H9C2 cells (undifferentiated rat cardiomyoblasts) were transfected with Hdac1/2/3 specific small interfering RNA (siRNA). Based on the results of siRNA transfection, cells were transfected with Hdac3 expression plasmid and subsequently mock-treated or treated with 8.0 mmol/L VPA. Hdac1/2/3 as well as Vangl2/Scrib/Rac1 mRNA and protein levels were determined by real-time quantitative polymerase chain reaction and Western blotting, respectively. Total HDAC activity was detected by colorimetric assay. Results: VPA could induce CHD (P < 0.001) and inhibit mRNA or protein expression of Hdac1/2/3 as well as Vangl2/Scrib in fetal hearts, in association with total Hdac activity repression (all P < 0.05). In vitro, Hdac3 inhibition could significantly decrease Vangl2/Scrib expression (P < 0.01), while knockdown of Hdac1/2 had no influence (P > 0.05); VPA exposure dramatically decreased the expression of Vanlg2/Scrib together with Hdac activity (P < 0.01), while overexpression of Hdac3 could rescue the VPA-induced inhibition (P > 0.05). Conclusion VPA could inhibit Hdac1/2/3, Vangl2/Scrib, or total Hdac activity both in vitro and in vivo and Hdac3 might participate in the process of VPA-induced cardiac developmental anomalies.
Animals ; Cell Polarity ; Enzyme Inhibitors ; adverse effects ; Female ; Fetal Heart ; embryology ; Heart Defects, Congenital ; chemically induced ; physiopathology ; Histone Deacetylase Inhibitors ; Histone Deacetylases ; drug effects ; physiology ; Mice ; Mice, Inbred C57BL ; Nerve Tissue Proteins ; Pregnancy ; Rats ; Transfection ; Valproic Acid ; adverse effects

Fetal heart sound is nonlinear and non-stationary, which contains a lot of noise when it is colleced, so the denoising method is important. We proposed a new denoising method in our study. Firstly, we chose the preprocessing of low-pass filter with a cutoff frequency of 200 Hz and the resampling. Secondly, we decomposed the signal based on empirical mode decomposition method (EMD) of Hilbert-Huang transform, then denoised some selected target components with wavelet soft threshold adaptive noise cancellation algorithm. Finally we got the clean fetal heart sound by combining the target components. In the EMD, we used a mask signal to eliminate the mode mixing problem, used mirroring extension method to eliminate the end effect, and referenced the stopping rule from the research of Rilling. This method eliminated the baseline drift and noise at once. To compare with wavelet transform (WT), mathematical morphology (MM) and the Fourier transform (FT), the SNR was improved obviously, and the RMSE was the minimum, which could satisfy the need of the practical application.
Algorithms ; Fetal Heart ; physiology ; Heart Sounds ; Humans ; Signal Processing, Computer-Assisted ; Signal-To-Noise Ratio ; Wavelet Analysis

OBJECTIVE: To evaluate the superiority, feasibility and clinical signifi cance of the four-dimensional spatio-temporal image correlation (STIC) in detection of the right ventricle function of fetus. METHODS: Th e STIC dynamic images of 180 normal fetal hearts at 24+0 to 37+6 weeks of gestation were obtained by the three-dimensional (3D) probe. Th e post-process evaluation was done off -line with the virtual organ computer-aided analysis (VOCAL) software. The computer recorded the right ventricular end-diastolic volume (RVEDV), right ventricular end-systolic volume (RVESV), and then calculated the right stroke volume (RSV), the right cardiac output (RCO) and the right ejection fraction (REF). RCO was standardized by biometric measurements obtained at prenatal screening, including head circumference (HC), abdominal circumference (AC), femur length (FL) and estimated fetal weight (EFW). RESULTS: The overall successful rate in image acquisition was 83.89% and the repeatability was favorable. After the standardization of fetal biometric parameters (HC, AC, FL) and the right ventricle function indexes (RVEDV, RVESV, RSV), RCO was increased with the gestational age while the REF and RCO/EFW fluctuated within a certain range. CONCLUSION STIC technique can accurately and objectively measure the fetal ventricular volume and it might be a potential strategy in the clinical assessment of the fetal cardiac function.
Biometry ; Female ; Fetal Heart ; diagnostic imaging ; physiology ; Gestational Age ; Humans ; Pregnancy ; Stroke Volume ; Ultrasonography, Prenatal ; Ventricular Function, Right

OBJECTIVETo study the effects of dopamine and phenylephrine for treatment of hypotension during cesarean section under combined spinal epidural anesthesia (CSEA) on the stereology of the placenta.

METHODSForty puerperants undergoing cesarean section under CSEA were randomly divided into dopamine group and phenylephrine group. Ropivacaine (16 mg) was administered immediately after spinal anethesia. Blood pressure was maintained near the baseline by adjusting the drug infusion rate. Fetal blood gas, Apgar score, and placental villus microvascular stereological changes were observed during the operation.

RESULTSThe microvascular density was significantly lower in dopamine group than in phenylephrine group (P<0.05). Phenylephrine group showed significantly lower umbilical artery blood pH than dopamine group (P<0.05). The Apgar score and blood pressure were comparable between the two groups (P>0.05). Compared to the baseline, both of the two groups showed significantly lowered heart rate during the operation (P<0.01).

CONCLUSIONDopamine is associated with the risk of fetal acidosis. Phenylephrine is helpful for preventing hypotension by increasing placental blood flow and improving oxygen supply to ensure maternal and fetal safety during cesarean section.

Amides ; administration & dosage ; Anesthesia, Spinal ; Apgar Score ; Blood Gas Analysis ; Blood Pressure ; Cesarean Section ; Dopamine ; administration & dosage ; Female ; Fetal Blood ; Fetus ; Heart Rate ; Humans ; Hypotension ; drug therapy ; Infant, Newborn ; Oxygen ; Phenylephrine ; administration & dosage ; Placenta ; drug effects ; physiology ; Pregnancy ; Vasoconstrictor Agents ; administration & dosage

Independent component analysis (ICA) aims at decomposing an observed random vector into statistically independent variables. Fast independent component analysis (FastICA) algorithm and its variants are catching more attention because of their simplicity and convergence speed. In this paper, a novel method referred to robust independent component analysis (RobustICA), based on normalized kurtosis and optimal step-size, is analyzed in detail. When applied to fetal electrocardiogram (FECG) extraction and compared with FastICA, it gave decent results and showed prosperous future usages.
Algorithms ; Electrocardiography ; Female ; Fetal Heart ; physiology ; Fetal Monitoring ; Fetus ; Humans ; Pregnancy ; Principal Component Analysis ; Signal Processing, Computer-Assisted

The purpose of this study was to investigate the feasibility of different fetal heart rate (FHR) ranges in the nonstress test (NST) and to better understand the meaning of mild bradycardia and/or tachycardia without non-reassuring patterns. We employed the heredity to show that mild bradycardia (100-119 beats per minute, bpm) and mild tachycardia (161-180 bpm) regressed to the normal FHR range (120-160 bpm). We used linear regression to analyze FHR data from FHR tracings recorded 10 min before (NST, as the predictor) and 10 min after vibroacoustic stimulation testing (as the dependent variable). Acceleration for 15 bpm-15 seconds (Acc1515) and deceleration for 15 bpm-15 seconds (Dec1515) in the NST were also analyzed for each group. The slope of the best-fit line was the largest in the mild bradycardia group and the smallest in the normal range group. Dec1515 was most prominent in mild tachycardia and both the mild bradycardia and tachycardia groups regressed towards the mean FHR range. Therefore, we propose that both mild bradycardia and tachycardia of FHR in non-acute situations (range between 100 and 180 bpm) are not regarded a pathologic signal for clinical use.
Acoustic Stimulation ; Bradycardia/physiopathology ; Female ; *Fetal Monitoring ; Heart Rate, Fetal/*physiology ; Humans ; Pregnancy ; Pregnancy Trimester, Third ; Regression Analysis ; Tachycardia/physiopathology

Fetal electrocardiogram (FECG) is an objective index of the activities of fetal cardiac electrophysiology. The acquired FECG is interfered by maternal electrocardiogram (MECG). How to extract the fetus ECG quickly and effectively has become an important research topic. During the non-invasive FECG extraction algorithms, independent component analysis(ICA) algorithm is considered as the best method, but the existing algorithms of obtaining the decomposition of the convergence properties of the matrix do not work effectively. Quantum particle swarm optimization (QPSO) is an intelligent optimization algorithm converging in the global. In order to extract the FECG signal effectively and quickly, we propose a method combining ICA and QPSO. The results show that this approach can extract the useful signal more clearly and accurately than other non-invasive methods.
Algorithms ; Electrocardiography ; methods ; Fetal Heart ; physiology ; Humans ; Principal Component Analysis ; Quantum Theory ; Signal Processing, Computer-Assisted

According to the independent component analysis (ICA) theory, in order to seperate the fetal electrocardiograms (FECG) from the observed data, we must use the larger number of observed signals than that of independent components. This requires that the number of the sentors on the electrocardiograph must be larger than a particular value, which is in practice hard to be satisfied. So we proposed another algorithm using fewer channels of abdominal electrocardiograms, which combines ICA and adaptive noise cancellation to extract the FECG from two leads of abdominal electrocardiograms. The experiment results showed that we could obtain clear FECG by the method we proposed.
Algorithms ; Electrocardiography ; methods ; Female ; Fetal Heart ; physiology ; Humans ; Pregnancy ; Principal Component Analysis ; Signal Processing, Computer-Assisted

The present paper is a research on independent component analysis (ICA) method in the fetal electrocardiogram (FECG) extraction. Based on the fundamental model for the ICA and the fixed-point FastICA algorithm using negentropy, damped Newton iteration was used in place of Newton iteration. The algorithm was improved in order to overcome the drawbacks where it is more sensitive to choosing the initial value. The improved algorithm was used to extract the FECG. A synthetic ECG was used in the experiments, and three simulation signal sources were selected, including two sources of ECG and one Gaussian noise source. The experimental results were satisfactory, The convergence rate was faster and the error was smaller.
Algorithms ; Electrocardiography ; methods ; Female ; Fetal Heart ; physiology ; Humans ; Pregnancy ; Principal Component Analysis ; methods ; Signal Processing, Computer-Assisted