暂无摘要。

OBJECTIVE: Obstetricians use Cardiotocography (CTG), which is the continuous recording of fetal heart rate and uterine contraction, to assess fetal health status. Deep learning models for intelligent fetal monitoring trained on extensively labeled and identically distributed CTG records have achieved excellent performance. However, creation of these training sets requires excessive time and specialist labor for the collection and annotation of CTG signals. Previous research has demonstrated that multicenter studies can improve model performance. However, models trained on cross-domain data may not generalize well to target domains due to variance in distribution among datasets. Hence, this paper conducted a multicenter study with Deep Semi-Supervised Domain Adaptation (DSSDA) for intelligent interpretation of antenatal CTG signals. This approach helps to align cross-domain distribution and transfer knowledge from a label-rich source domain to a label-scarce target domain.
METHODS: We proposed a DSSDA framework that integrated Minimax Entropy and Domain Invariance (DSSDA-MMEDI) to reduce inter-domain gaps and thus achieve domain invariance. The networks were developed using GoogLeNet to extract features from CTG signals, with fully connected, softmax layers for classification. We designed a Dynamic Gradient-driven strategy based on Mutual Information (DGMI) to unify the losses from Minimax Entropy (MME), Domain Invariance (DI), and supervised cross-entropy during iterative learning.
RESULTS: We validated our DSSDA model on two datasets collected from collaborating healthcare institutions and mobile terminals as the source and target domains, which contained 16,355 and 3,351 CTG signals, respectively. Compared to the results achieved with deep learning networks without DSSDA, DSSDA-MMEDI significantly improved sensitivity and F1-score by over 6%. DSSDA-MMEDI also outperformed other state-of-the-art DSSDA approaches for CTG signal interpretation. Ablation studies were performed to determine the unique contribution of each component in our DSSDA mechanism.
CONCLUSIONS: The proposed DSSDA-MMEDI is feasible and effective for alignment of cross-domain data and automated interpretation of multicentric antenatal CTG signals with minimal annotation cost.

BACKGROUND: In clinical medicine, fetal heart rate (FHR) monitoring using cardiotocography (CTG) is one of the most commonly used methods for assessing fetal acidosis. However, as the visual interpretation of CTG depends on the subjective judgment of the clinician, this has led to high inter-observer and intra-observer variability, making it necessary to introduce automated diagnostic techniques.
METHODS: In this study, we propose a computer-aided diagnostic algorithm (Hybrid-FHR) for fetal acidosis to assist physicians in making objective decisions and taking timely interventions. Hybrid-FHR uses multi-modal features, including one-dimensional FHR signals and three types of expert features designed based on prior knowledge (morphological time domain, frequency domain, and nonlinear). To extract the spatiotemporal feature representation of one-dimensional FHR signals, we designed a multi-scale squeeze and excitation temporal convolutional network (SE-TCN) backbone model based on dilated causal convolution, which can effectively capture the long-term dependence of FHR signals by expanding the receptive field of each layer\'s convolution kernel while maintaining a relatively small parameter size. In addition, we proposed a cross-modal feature fusion (CMFF) method that uses multi-head attention mechanisms to explore the relationships between different modalities, obtaining more informative feature representations and improving diagnostic accuracy.
RESULTS: Our ablation experiments show that the Hybrid-FHR outperforms traditional previous methods, with average accuracy, specificity, sensitivity, precision, and F1 score of 96.8, 97.5, 96, 97.5, and 96.7%, respectively.
CONCLUSIONS: Our algorithm enables automated CTG analysis, assisting healthcare professionals in the early identification of fetal acidosis and the prompt implementation of interventions.

Intelligent cardiotocography (CTG) classification can assist obstetricians in evaluating fetal health. However, high classification performance is often achieved by complex machine learning (ML)-based models, which causes interpretability concerns. The trade-off between accuracy and interpretability makes it challenging for most existing ML-based CTG classification models to popularize in prenatal clinical applications.
Aiming to improve CTG classification performance and prediction interpretability, a hybrid model was proposed using a stacked ensemble strategy with mixed features and Kernel SHapley Additive exPlanations (SHAP) framework. Firstly, the stacked ensemble classifier was established by employing support vector machines (SVM), extreme gradient boosting (XGB), and random forests (RF) as base learners, and backpropagation (BP) as a meta learner whose input was mixed with the CTG features and the probability value of each category output by base learners. Then, the public and private CTG datasets were used to verify the discriminative performance. Furthermore, Kernel SHAP was applied to estimate the contribution values of features and their relationships to the fetal states.
For intelligent CTG classification using 10-fold cross-validation, the accuracy and average F1 score were 0.9539 and 0.9249 in the public dataset, respectively; and those were 0.9201 and 0.8926 in the private dataset, respectively. For interpretability, the explanation results indicated that accelerations (AC) and the percentage of time with abnormal short-term variability (ASTV) were the key determinants. Specifically, the probability of abnormality increased and that of the normal state decreased as the value of ASTV grew. In addition, the likelihood of the normal status rose with the increase of AC.
The proposed model has high classification performance and reasonable interpretability for intelligent fetal monitoring.

OBJECTIVE: To propose a computerized system utilizing multiscene analysis based on a support vector machine (SVM) and convolutional neural network (CNN) to assess cardiotocography (CTG) intelligently.
METHODS: We retrospectively collected 2542 CTG records of singleton pregnancies delivered at the maternity ward of the First Affiliated Hospital of Xi\'an Jiaotong University from October 10, 2020, to August 7, 2021. CTG records were divided into five categories (baseline, variability, acceleration, deceleration, and normality). Apart from the category of normality, the other four different categories of abnormal data correspond to four scenes. Each scene was divided into training and testing sets at 9:1 or 7:3. We used three computer algorithms (dynamic threshold, SVM, and CNN) to learn and optimize the system. Accuracy, sensitivity, and specificity were performed to evaluate performance.
RESULTS: The global accuracy, sensitivity, and specificity of the system were 93.88%, 93.06%, and 94.33%, respectively. In acceleration and deceleration scenes, when the convolution kernel was 3, the test data set reached the highest performance.
CONCLUSIONS: The multiscene research model using SVM and CNN is a potential effective tool to assist obstetricians in classifying CTG intelligently.

OBJECTIVE: To investigate the association and the potential value of prelabour fetal heart rate short-term variability (STV) determined by computerised cardiotocography (cCTG) and maternal and fetal Doppler in predicting labour outcomes.
METHODS: Prospective cohort study.
METHODS: The Prince of Wales Hospital, a tertiary maternity unit, in Hong Kong SAR.
METHODS: Women with a term singleton pregnancy in latent phase of labour or before labour induction were recruited during May 2019-November 2021.
METHODS: Prelabour ultrasonographic assessment of fetal growth, Doppler velocimetry and prelabour cCTG monitoring including Dawes-Redman CTG analysis were registered shortly before induction of labour or during the latent phase of spontaneous labour.
METHODS: Umbilical cord arterial pH, emergency delivery due to pathological CTG during labour and neonatal intensive care unit (NICU)/special care baby unit (SCBU) admission.
RESULTS: Of the 470 pregnant women invited to participate in the study, 440 women provided informed consent and a total of 400 participants were included for further analysis. Thirty-four (8.5%) participants underwent emergency delivery for pathological CTG during labour. A total of 6 (1.50%) and 148 (37.00%) newborns required NICU and SCBU admission, respectively. Middle cerebral artery pulsatility index (MCA-PI) and MCA-PI z-score were significantly lower in pregnancies that required emergency delivery for pathological CTG during labour compared with those that did not (1.23 [1.07-1.40] versus 1.40 [1.22-1.64], p = 0.002; and 0.55 ± 1.07 vs. 0.12 ± 1.06), p = 0.049]. This study demonstrated a weakly positive correlation between umbilical cord arterial pH and prelabour log10 STV (r = 0.107, p = 0.035) and the regression analyses revealed that the contributing factors for umbilical cord arterial pH were smoking (p = 0.006) and prelabour log10 STV (p = 0.025).
CONCLUSIONS: In pregnant women admitted in latent phase of labour or for induction of labour at term, prelabour cCTG STV had a weakly positive association with umbilical cord arterial pH but was not predictive of emergency delivery due to pathological CTG during labour.

Any acute and profound reduction in fetal oxygenation increases the risk of anaerobic metabolism in the fetal myocardium and, hence, the risk of lactic acidosis. On the contrary, in a gradually evolving hypoxic stress, there is sufficient time to mount a catecholamine-mediated increase in the fetal heart rate to increase the cardiac output and redistribute oxygenated blood to maintain an aerobic metabolism in the fetal central organs. When the hypoxic stress is sudden, profound, and sustained, it is not possible to continue to maintain central organ perfusion by peripheral vasoconstriction and centralization. In case of acute deprivation of oxygen, the immediate chemoreflex response via the vagus nerve helps reduce fetal myocardial workload by a sudden drop of the baseline fetal heart rate. If this drop in the fetal heart rate continues for >2 minutes (American College of Obstetricians and Gynecologists\' guideline) or 3 minutes (National Institute for Health and Care Excellence or physiological guideline), it is termed a prolonged deceleration, which occurs because of myocardial hypoxia, after the initial chemoreflex. The revised International Federation of Gynecology and Obstetrics guideline (2015) considers the prolonged deceleration to be a \"pathologic\" feature after 5 minutes. Acute intrapartum accidents (placental abruption, umbilical cord prolapse, and uterine rupture) should be excluded immediately, and if they are present, an urgent birth should be accomplished. If a reversible cause is found (maternal hypotension, uterine hypertonus or hyperstimulation, and sustained umbilical cord compression), immediate conservative measures (also called intrauterine fetal resuscitation) should be undertaken to reverse the underlying cause. In reversible causes of acute hypoxia, if the fetal heart rate variability is normal before the onset of deceleration, and normal within the first 3 minutes of the prolonged deceleration, then there is an increased likelihood of recovery of the fetal heart rate to its antecedent baseline within 9 minutes with the reversal of the underlying cause of acute and profound reduction in fetal oxygenation. The continuation of the prolonged deceleration for >10 minutes is termed \"terminal bradycardia,\" and this increases the risk of hypoxic-ischemic injury to the deep gray matter of the brain (the thalami and the basal ganglia), predisposing to dyskinetic cerebral palsy. Therefore, any acute fetal hypoxia, which manifests as a prolonged deceleration on the fetal heart rate tracing, should be considered an intrapartum emergency requiring an immediate intervention to optimize perinatal outcome. In uterine hypertonus or hyperstimulation, if the prolonged deceleration persists despite stopping the uterotonic agent, then acute tocolysis is recommended to rapidly restore fetal oxygenation. Regular clinical audit of the management of acute hypoxia, including the \"the onset of bradycardia to delivery interval,\" may help identify organizational and system issues, which may contribute to poor perinatal outcomes.

The onset of regular, strong, and progressive uterine contractions may result in both mechanical (compression of the fetal head and/or umbilical cord) and hypoxic (repetitive and sustained compression of the umbilical cord or reduction in uteroplacental oxygenation) stresses to a human fetus. Most fetuses are able to mount effective compensatory responses to avoid hypoxic-ischemic encephalopathy and perinatal death secondary to the onset of anaerobic metabolism within the myocardium, culminating in myocardial lactic acidosis. In addition, the presence of fetal hemoglobin, which has a higher affinity for oxygen even at low partial pressures of oxygen than the adult hemoglobin, especially increased amounts of fetal hemoglobin (ie, 180-220 g/L in fetuses vs 110-140 g/L in adults), helps the fetus to withstand hypoxic stresses during labor. Different national and international guidelines are currently being used for intrapartum fetal heart rate interpretation. These traditional classification systems for fetal heart rate interpretation during labor are based on grouping certain features of fetal heart rate (ie, baseline fetal heart rate, baseline variability, accelerations, and decelerations) into different categories (eg, category I, II, and III tracings, \"normal, suspicious, and pathologic\" or \"normal, intermediary, and abnormal\"). These guidelines differ from each other because of the features included within different categories and because of their arbitrary time limits stipulated for each feature to warrant an obstetrical intervention. This approach fails to individualize care because the \"ranges of normality\" for stipulated parameters apply to the population of human fetuses and not to the individual fetus in question. Moreover, different fetuses have different reserves and compensatory responses and different intrauterine environments (presence of meconium staining of amniotic fluid, intrauterine inflammation, and the nature of uterine activity). Pathophysiological interpretation of fetal heart rate tracing is based on the application of the knowledge of fetal responses to intrapartum mechanical and/or hypoxic stress in clinical practice. Both experimental animal studies and observational human studies suggest that, just like adults undertaking a treadmill exercise, human fetuses show predictable compensatory responses to a progressively evolving intrapartum hypoxic stress. These responses include the onset of decelerations to reduce myocardial workload and preserve aerobic metabolism, loss of accelerations to abolish nonessential somatic body movements, and catecholamine-mediated increases in the baseline fetal heart rate and effective redistribution and centralization to protect the fetal central organs (ie, the heart, brain, and adrenal glands), which are essential for intrauterine survival. Moreover, it is essential to incorporate the clinical context (progress of labor, fetal size and reserves, presence of meconium staining of amniotic fluid and intrauterine inflammation, and fetal anemia) and understand the features suggestive of fetal compromise in nonhypoxic pathways (eg, chorioamnionitis and fetomaternal hemorrhage). It is important to appreciate that the timely recognition of the speed of onset of intrapartum hypoxia (ie, acute, subacute, and gradually evolving) and preexisting uteroplacental insufficiency (ie, chronic hypoxia) on fetal heart rate tracing is crucial to improve perinatal outcomes.

CTG (Cardiotocography) is an effective tool for fetal status assessment. Clinically, doctors mainly evaluate the health of fetus by observing FHR (fetal heart rate). The rapid development of Artificial Intelligence has led realization of computer-aided CTG technology, Intelligent CTG classification based on FHR is a fundamental component of these technologies. Its implementation can provide doctors with auxiliary decisions. Most of existing FHR classification methods are based on combing different deep learning models, such as CNN (Convolutional Neural Network), LSTM (Long short-term memory) and Transformer. However, these studies ignore the balance of positive and negative samples in dataset and the matching degree between model and FHR classification task, which reduces the classification accuracy. In this paper, we mainly discuss two major problems in previous FHR classification studies: reduce class imbalance and select appropriate convolution kernel. To address above two problems, we propose a data augmentation method based on ECMN (Edge Clipping and Multiscale Noise) to resolve class imbalance. Subsequently, we introduce a one-dimensional long convolutional layer, which use trend area to calculate the appropriate convolution kernel. Based on appropriate convolution kernel, an improved residual structure with attention mechanism named TGLCN (Trend-Guided Long Convolution Network) is proposed to improve FHR classification accuracy. Finally, horizontal and longitudinal experiments show that the TGLCN obtains high classification accuracy and speed of parameter adjustment.

OBJECTIVE: Prenatal fetal monitoring, which can monitor the growth and health of the fetus, is very vital for pregnant women before delivery. During pregnancy, it is crucial to judge whether the fetus is abnormal, which helps obstetricians carry out early intervention to avoid fetal hypoxia and even death. At present, clinical fetal monitoring widely used fetal heart rate monitoring equipment. Fetal heart rate and uterine contraction signals obtained by fetal heart monitoring equipment are important information to evaluate fetal health status.
METHODS: This paper is based on 1D-CNN (One Dimension Convolutional Neural Network) and GRU (Gate Recurrent Unit). We preprocess the obtained data and enhances them, to make the proportion of number of instances in different class in the training set is same.
RESULTS: In model performance evaluation, standard evaluation indicators are used, such as accuracy, sensitivity, specificity, and ROC (receiver operating characteristic). Finally, the accuracy of our model in the test set is 95.15%, the sensitivity is 96.20%, and the specificity is 94.09%.
CONCLUSIONS: In fetal heart rate monitoring, this paper proposes a 1D-CNN and bidirectional GRU hybrid models, and the fetal heart rate and uterine contraction signals given by monitoring are used as input feature to classify the fetal health status. The results show that our approach is effective in evaluating fetal health status and can assists obstetricians in clinical decision-making. And provide a baseline for the introduction of 1D-CNN and bidirectional GRU hybrid models into the evaluation of fetal health status.