Fetal electrocardiogram ST-segment analysis and prediction of neonatal acidosis

Fetal monitoring technologies for the detection of intrapartum hypoxia - challenges and opportunities

Intrapartum fetal hypoxia is related to long-term morbidity and mortality of the fetus and the mother. Fetal surveillance is extremely important to minimize the adverse outcomes arising from fetal hypoxia during labour. Several methods have been used in current clinical practice to monitor fetal well-being. For instance, biophysical technologies including cardiotocography, ST-analysis adjunct to cardiotocography, and Doppler ultrasound are used for intrapartum fetal monitoring. However, these technologies result in a high false-positive rate and increased obstetric interventions during labour. Alternatively, biochemical-based technologies including fetal scalp blood sampling and fetal pulse oximetry are used to identify metabolic acidosis and oxygen deprivation resulting from fetal hypoxia. These technologies neither improve clinical outcomes nor reduce unnecessary interventions during labour. Also, there is a need to link the physiological changes during fetal hypoxia to fetal monitoring technologies. The objective of this article is to assess the clinical background of fetal hypoxia and to review existing monitoring technologies for the detection and monitoring of fetal hypoxia. A comprehensive review has been made to predict fetal hypoxia using computational and machine-learning algorithms. The detection of more specific biomarkers or new sensing technologies is also reviewed which may help in the enhancement of the reliability of continuous fetal monitoring and may result in the accurate detection of intrapartum fetal hypoxia.

A lightweight fetal distress-assisted diagnosis model based on a cross-channel interactive attention mechanism

Fetal distress is a symptom of fetal intrauterine hypoxia, which is seriously harmful to both the fetus and the pregnant woman. The current primary clinical tool for the assessment of fetal distress is Cardiotocography (CTG). Due to subjective variability, physicians often interpret CTG results inconsistently, hence the need to develop an auxiliary diagnostic system for fetal distress. Although the deep learning-based fetal distress-assisted diagnosis model has a high classification accuracy, the model not only has a large number of parameters but also requires a large number of computational resources, which is difficult to deploy to practical end-use scenarios. Therefore, this paper proposes a lightweight fetal distress-assisted diagnosis network, LW-FHRNet, based on a cross-channel interactive attention mechanism. The wavelet packet decomposition technique is used to convert the one-dimensional fetal heart rate (FHR) signal into a two-dimensional wavelet packet coefficient matrix map as the network input layer to fully obtain the feature information of the FHR signal. With ShuffleNet-v2 as the core, a local cross-channel interactive attention mechanism is introduced to enhance the model's ability to extract features and achieve effective fusion of multichannel features without dimensionality reduction. In this paper, the publicly available database CTU-UHB is used for the network performance evaluation. LW-FHRNet achieves 95.24% accuracy, which meets or exceeds the classification results of deep learning-based models. Additionally, the number of model parameters is reduced many times compared with the deep learning model, and the size of the model parameters is only 0.33 M. The results show that the lightweight model proposed in this paper can effectively aid in fetal distress diagnosis.

Optimizing oxygen therapy for preterm infants at birth: Are we there yet?

Premature infants undergo a complex postnatal adaptation at birth. For last two centuries, oxygen has been integral to respiratory support of preterm infants at birth. Excess oxygen can cause oxidative stress and tissue injury. Preterm infants due to lung immaturity may need oxygen for successful transition at birth. Although, considerable progress has been made in the last 3 decades, optimum oxygen therapy for preterm delivery room resuscitation remains unknown. In this review, we discuss the history and physiology behind oxygen therapy in the delivery room, evaluate current literature, provide practice points and point out knowledge gaps of oxygen therapy in preterm infant at birth.

DeepFHR: intelligent prediction of fetal Acidemia using fetal heart rate signals based on convolutional neural network

Background

Fetal heart rate (FHR) monitoring is a screening tool used by obstetricians to evaluate the fetal state. Because of the complexity and non-linearity, a visual interpretation of FHR signals using common guidelines usually results in significant subjective inter-observer and intra-observer variability. Objective: Therefore, computer aided diagnosis (CAD) systems based on advanced artificial intelligence (AI) technology have recently been developed to assist obstetricians in making objective medical decisions.

Methods

In this work, we present an 8-layer deep convolutional neural network (CNN) framework to automatically predict fetal acidemia. After signal preprocessing, the input 2-dimensional (2D) images are obtained using the continuous wavelet transform (CWT), which provides a better way to observe and capture the hidden characteristic information of the FHR signals in both the time and frequency domains. Unlike the conventional machine learning (ML) approaches, this work does not require the execution of complex feature engineering, i.e., feature extraction and selection. In fact, 2D CNN model can self-learn useful features from the input data with the prerequisite of not losing informative features, representing the tremendous advantage of deep learning (DL) over ML.

Results

Based on the test open-access database (CTU-UHB), after comprehensive experimentation, we achieved better classification performance using the optimal CNN configuration compared to other state-of-the-art methods: the averaged ten-fold cross-validation of the accuracy, sensitivity, specificity, quality index defined as the geometric mean of the sensitivity and specificity, and the area under the curve yielded results of 98.34, 98.22, 94.87, 96.53 and 97.82%, respectively

Conclusions

Once the proposed CNN model is successfully trained, the corresponding CAD system can be served as an effective tool to predict fetal asphyxia objectively and accurately.

Fetal electrocardiography ST-segment analysis for intrapartum monitoring: a critical appraisal of conflicting evidence and a way forward

BACKGROUND

In the past century, some areas of obstetric including intrapartum care have been slow to benefit from the dramatic advances in technology and medical care. Although fetal heart rate monitoring (cardiotocography) became available a half century ago, its interpretation often differs between institutions and countries, its diagnostic accuracy needs improvement, and a technology to help reduce the unnecessary obstetric interventions that have accompanied the cardiotocography is urgently needed.

STUDY DESIGN

During the second half of the 20th century, key findings in animal experiments captured the close relationship between myocardial glycogenolysis, myocardial workload, and ST changes, thus demonstrating that ST waveform analysis of the fetal electrocardiogram can provide information on oxygenation of the fetal myocardium and establishing the physiological basis for the use of electrocardiogram in intrapartum fetal surveillance.

RESULTS

Six randomized controlled trials, 10 meta-analyses, and more than 20 observational studies have evaluated the technology developed based on this principle. Nonetheless, despite this intensive assessment, differences in study protocols, inclusion criteria, enrollment rates, clinical guidelines, use of fetal blood sampling, and definitions of key outcome parameters, as well as inconsistencies in randomized controlled trial data handling and statistical methodology, have made this voluminous evidence difficult to interpret. Enormous resources spent on randomized controlled trials have failed to guarantee the generalizability of their results to other settings or their ability to reflect everyday clinical practice.

CONCLUSION

The latest meta-analysis used revised data from primary randomized controlled trials and data from the largest randomized controlled trials from the United States to demonstrate a significant reduction of metabolic acidosis rates by 36% (odds ratio, 0.64; 95% confidence interval, 0.46-0.88) and operative vaginal delivery rates by 8% (relative risk, 0.92; 95% confidence interval, 0.86-0.99), compared with cardiotocography alone.

Oxygen Therapy in the Delivery Room: What Is the Right Dose?

Oxygen is the most commonly used medicine used during neonatal resuscitation in the delivery room. Oxygen therapy in delivery room should be used judiciously to avoid oxygen toxicity while delivering sufficient oxygen to prevent hypoxia. Measurement of appropriate oxygenation relies on pulse oximetry, but adequate ventilation and perfusion are equally important for oxygen delivery. In this article, we review oxygenation while transitioning from fetal to neonatal life, the importance of appropriate oxygen therapy, its measurement in the delivery room, and current recommendations for oxygen therapy and its limitations.

Evaluation of Abdominal Fetal Electrocardiography in Early Intrauterine Growth Restriction

Objectives: This descriptive study was performed to evaluate the capability of a non-invasive transabdominal electrocardiographic system to extract clear fetal electrocardiographic (FECG) measurements from intrauterine growth restricted (IUGR) fetuses and to assess whether abdominal FECG parameters can be developed as markers for evaluating the fetal cardiac status in IUGR.

Methods: Transabdominal FECG was attempted in 20 controls and 15 IUGR singleton pregnancies at 20+0−33+6 weeks gestation. Standard ECG parameters were compared between the study groups and evaluated for their correlation. Accuracy for the prediction of IUGR by cut off values of the different FECG parameters was also determined.

Results: Clear P-QRST complexes were recognized in all cases. In the IUGR fetuses, the QT and QTc intervals were significantly prolonged (p = 0.017 and p = 0.002, respectively). There was no correlation between ECG parameters and Doppler or other indices to predict IUGR. The generation of cut off values for detecting IUGR showed increasing sensitivities but decreasing specificities with the prolongation of ECG parameters.

Conclusion: The study of fetal electrocardiophysiology is now feasible through a non-invasive transabdominal route. This study confirms the potential of FECG as a clinical screening tool to aid diagnosis and management of fetuses after key limitations are addressed. In the case of IUGR, both QT and QTc intervals were significantly prolonged and thus validate earlier study findings where both these parameters were found to be markers of diastolic dysfunction. This research is a useful prelude to a test of accuracy and Receiver Operating Characteristics (ROC) study.

Intrapartum cardiotocography (CTG) and ST-analysis of labor in diabetic patients

AIM

To determine the prevalence and types of intrapartum cardiotocography (CTG) patterns and investigate their relationship to moderate acidemia in term fetuses of diabetic mothers. Also, to assess if the combination of fetal electrocardiogram (FECG) and those CTG patterns strengthens the association with moderate acidemia.

MATERIAL AND METHODS

The material for this study is obtained from the Swedish randomized control trial and the European Union ST-analysis trial. We developed an analytical model for CTG patterns based on the progress in CTG changes, in a longitudinal periodic manner. The model was then combined with information regarding changes in ST interval that indicate threatening asphyxia, and the findings were analyzed to determine correlation with the presence of moderate acidemia at birth.

RESULTS

This study involved data of 413 diabetic mothers. A preterminal CTG was more common in the diabetes mellitus (DM) group (6/70, 8.6%) than in the gestational diabetes (GD) group (3/307, 1.0%; P=0.003). For diabetic mothers (i.e., DM+GD) with a normal CTG at the start of monitoring, the presence of FECG data indicating asphyxia significantly increased the likelihood of an umbilical artery pH<7.15 at birth [odds ratio (OR)=3.65, 95% confidence interval (CI)=1.33-10.05]. Among labors where the CTG was non-reassuring at the start of monitoring, no significant association was found between pH<7.15 and indication to intervene according to FECG information (OR=1.51, 95% CI=0.33-7.0).

CONCLUSIONS

A preterminal CTG is more common in the fetuses of DM than GD mothers during labor. When CTG was normal at the start of recording, the addition of FECG information gave a significant add on information to predict moderate acidemia.

ST segment analysis as an adjunct to electronic fetal monitoring, Part I: background, physiology, and interpretation

Fetal electrocardiogram (ECG) ST segment analysis (STAN) was approved in 2005 in the United States as an adjunct to electronic fetal heart rate monitoring to determine whether obstetrical intervention is warranted when there is an increased risk for developing metabolic acidosis. STAN has utility in the reduction of fetal acidosis at birth, decreased need for fetal scalp blood sampling during labor, and decreased need for operative vaginal delivery and emergency cesarean delivery for fetal indications. This article discusses specific fetal ECG changes and their significance and the use of the STAN system.

ST segment analysis (STAN) as an adjunct to electronic fetal monitoring, Part II: clinical studies and future directions

The Part 1 article focused on the physiology and general system for interpretation of ST segment analysis (STAN) systems. This article focuses on prior clinical studies of STAN and future research directions.

ST analysis as an adjunct to electronic fetal monitoring: an overview

Conventional electronic fetal monitoring in the United States has value as a screening tool but is extremely limited as a diagnostic tool. ST analysis was developed as an adjunctive technology, able to measure changes in the ST segment of the fetal electrocardiogram during periods of hypoxia, improving the identification of the fetus at risk for metabolic acidemia at birth. Currently used only in a handful of hospitals in the United States, studies abroad have demonstrated that an integrated approach utilizing electronic fetal monitoring, ST analysis, and standardized guidelines in a selected patient population can improve neonatal outcome, decrease acidemia at birth, and decrease obstetric operative delivery. Research is needed to determine whether similar results are possible in the US population.

Cesarean section for suspected fetal distress, continuous fetal heart monitoring and decision to delivery time

OBJECTIVE

To find out the efficacy of continuous fetal heart monitoring by analyzing the cases of cesarean section for nonreassuring fetal heart in labor, detected by cardiotocography (CTG) and correlating these cases with perinatal outcome. To evaluate whether a 30 minute decision to delivery (D-D) interval for emergency cesarean section influences perinatal outcome.

METHODS

This was a prospective observational study of 217 patients who underwent cesarean section at > or = 36 weeks for non-reassuring fetal heart in labor detected by CTG. The maternal demographic profile, specific types of abnormal fetal heart rate tracing and the decision to delivery time interval were noted. The adverse immediate neonatal outcomes in terms of Apgar score <7 at 5 minutes, umbilical cord thornH <7.10, neonates requiring immediate ventilation and NICU admissions were recorded. The correlation between non-reassuring fetal heart, decision to delivery interval and neonatal outcome were analyzed.

RESULTS

Out of 3148 patients delivered at > or = 36 weeks, 217 (6.8%) patients underwent cesarean section during labor primarily for non-reassuring fetal heart. The most common fetal heart abnormality was persistent bradycardia in 106 (48.8%) cases followed by late deceleration in 38 (17.5%) cases and decreased beat to beat variability in 17 (7.8%) cases. In 33 (15.2%) babies the 5 minutes Apgar score was <7 out of which 13 (5.9%) babies had cord thornH <7.10. Thirty three (15.2%) babies required NICU admission for suspected birth asphyxia. Rest 184 (84.7%) neonates were born healthy and cared for by mother. Regarding decision to delivery interval of < or =30 minutes versus >30 minutes, there was no significant difference in the incidence of Apgar score <7 at 5 minutes, cord pH <7.10 and new born babies requiring immediate ventilation. But the need for admission to NICU in the group of D-D interval < or = 30 minutes was significantly higher compared to the other group where D-D interval was >30 minutes.

CONCLUSION

Non-reassuring fetal heart rate detected by CTG did not correlate well with adverse neonatal outcome. There was no significant difference in immediate adverse neonatal outcome whether the D-D time interval was < or = 30 minutes or >30 minutes; contrary to this, NICU admission for suspected birth asphyxia in </= 30 minutes group was significantly higher.

A French randomized controlled trial of ST-segment analysis in a population with abnormal cardiotocograms during labor

OBJECTIVE

The purpose of this study was to assess whether knowledge of ST-segment analysis was associated with a reduction in operative deliveries for nonreassuring fetal status (NRFS) or with a need for at least 1 scalp pH during labor.

STUDY DESIGN

Seven hundred ninety-nine women at term with abnormal cardiotocography or meconium-stained amniotic fluid (7%) were assigned randomly to the intervention group (cardiotocography + STAN) or the control group (cardiotocography) in 2 university hospitals in Strasbourg, France. Scalp pH testing was optional in both groups. Abnormal neonatal outcome was pH <7.05 or umbilical cord blood artery base deficit of >12 or a 5-min Apgar score of <7 or neonatal intensive care unit admission or convulsions or neonatal death. Study power was 80% for the detection of a prespecified reduction from 50%-40% in operative delivery for NRFS.

RESULTS

The operative delivery (cesarean or instrumental) rate for NRFS did not differ between the 2 groups: 33.6% (134/399) in the cardiotocography + STAN analysis group vs 37% (148/400) in the cardiotocography group (relative risk, 0.91; 95% CI, 0.75-1.10). The rate of operative delivery for dystocia was also similar in both groups. The percentage of women whose fetus had at least 1 scalp pH measurement during labor was substantially lower in the group with ST-segment analysis: 27% compared with 62% (relative risk, 0.44; 95% CI, 0.36-0.52). Neonatal outcomes did not differ significantly between groups.

CONCLUSION

In a population with abnormal cardiotocography in labor, cardiotocography combined with ST-segment analysis was not associated with a reduction in operative deliveries for NRFS. The proportion of infants without scalp pH sampling during labor increased substantially, however.