Prediction of preterm deliveries from EHG signals using machine learning

Multifractal Analysis of Uterine Electromyography Signals for the Assessment of Progression of Pregnancy in Term Conditions

OBJECTIVES

The objectives of this paper are to examine the source of multifractality in uterine electromyography (EMG) signals and to study the progression of pregnancy in the term (gestation period > 37 weeks) conditions using multifractal detrending moving average (MFDMA) algorithm.

METHODS

The signals for the study, considered from an online database, are obtained from the surface of abdomen during the second (T1) and third trimester (T2). The existence of multifractality is tested using Hurst and scaling exponents. With the intention of identifying the origin of multifractality, the preprocessed signals are converted to shuffle and surrogate data. The original and the transformed signals are subjected to MFDMA to extract multifractal spectrum features, namely strength of multifractality, maximum, minimum, and peak singularity exponents.

RESULTS

The Hurst and scaling exponents extracted from the signals indicate that uterine EMG signals are multifractal in nature. Further analysis shows that the source of multifractality is mainly owing to the presence of long-range correlation, which is computed as 79.98% in T1 and 82.43% in T2 groups. Among the extracted features, the peak singularity exponent and strength of multifractality show statistical significance in identifying the progression of pregnancy. The corresponding coefficients of variation are found to be low, which show that these features have low intersubject variability.

CONCLUSION

It appears that the multifractal analysis can help in investigating the progressive changes in uterine muscle contractions during pregnancy.

Uterine electromyography for discrimination of labor imminence in women with threatened preterm labor under tocolytic treatment

As one of the main aims of obstetrics is to be able to detect imminent delivery in patients with threatened preterm labor, the techniques currently used in clinical practice have serious limitations in this respect. The electrohysterogram (EHG) has now emerged as an alternative technique, providing relevant information about labor onset when recorded in controlled checkups without administration of tocolytic drugs. The studies published to date mainly focus on EHG-burst analysis and, to a lesser extent, on whole EHG window analysis. The study described here assessed the ability of EHG signals to discriminate imminent labor (< 7 days) in women with threatened preterm labor undergoing tocolytic therapy, using both EHG-burst and whole EHG window analyses, by calculating temporal, spectral, and non-linear parameters. Only two non-linear EHG-burst parameters and four whole EHG window analysis parameters were able to distinguish the women who delivered < 7 days from the others, showing that EHG can provide relevant information on the approach of labor, even in women with threatened preterm labor under the effects of tocolytic therapy. The whole EHG window outperformed the EHG-burst analysis and is seen as a step forward in the development of real-time EHG systems able to predict imminent labor in clinical praxis. Graphical abstract The ability of EHG recordings to predict imminent labor (< 7 days) was analyzed in preterm threatened patients undergoing tocolytic therapies by means of EHG-burst and whole EHG window analysis. The non-linear features were found to have better performance than the temporal and spectral parameters in separating women who delivered in less than 7 days from those who did not.

Characterization and automatic classification of preterm and term uterine records

Predicting preterm birth is uncertain, and numerous scientists are searching for non-invasive methods to improve its predictability. Current researches are based on the analysis of ElectroHysteroGram (EHG) records, which contain information about the electrophysiological properties of the uterine muscle and uterine contractions. Since pregnancy is a long process, we decided to also characterize, for the first time, non-contraction intervals (dummy intervals) of the uterine records, i.e., EHG signals accompanied by a simultaneously recorded external tocogram measuring mechanical uterine activity (TOCO signal). For this purpose, we developed a new set of uterine records, TPEHGT DS, containing preterm and term uterine records of pregnant women, and uterine records of non-pregnant women. We quantitatively characterized contraction intervals (contractions) and dummy intervals of the uterine records of the TPEHGT DS in terms of the normalized power spectra of the EHG and TOCO signals, and developed a new method for predicting preterm birth. The results on the characterization revealed that the peak amplitudes of the normalized power spectra of the EHG and TOCO signals of the contraction and dummy intervals in the frequency band 1.0-2.2 Hz, describing the electrical and mechanical activity of the uterus due to the maternal heart (maternal heart rate), are high only during term pregnancies, when the delivery is still far away; and they are low when the delivery is close. However, these peak amplitudes are also low during preterm pregnancies, when the delivery is still supposed to be far away (thus suggesting the danger of preterm birth); and they are also low or barely present for non-pregnant women. We propose the values of the peak amplitudes of the normalized power spectra due to the influence of the maternal heart, in an electro-mechanical sense, in the frequency band 1.0-2.2 Hz as a new biophysical marker for the preliminary, or early, assessment of the danger of preterm birth. The classification of preterm and term, contraction and dummy intervals of the TPEHGT DS, for the task of the automatic prediction of preterm birth, using sample entropy, the median frequency of the power spectra, and the peak amplitude of the normalized power spectra, revealed that the dummy intervals provide quite comparable and slightly higher classification performances than these features obtained from the contraction intervals. This result suggests a novel and simple clinical technique, not necessarily to seek contraction intervals but using the dummy intervals, for the early assessment of the danger of preterm birth. Using the publicly available TPEHG DB database to predict preterm birth in terms of classifying between preterm and term EHG records, the proposed method outperformed all currently existing methods. The achieved classification accuracy was 100% for early records, recorded around the 23rd week of pregnancy; and 96.33%, the area under the curve of 99.44%, for all records of the database. Since the proposed method is capable of using the dummy intervals with high classification accuracy, it is also suitable for clinical use very early during pregnancy, around the 23rd week of pregnancy, when contractions may or may not be present.

Artificial Intelligence: The Future of Obstetrics and Gynecology

Background

Artificial intelligence or 'big data' comprises of algorithms which aid in decision making. It has made an impact on a number of professions including obstetrics and gynecology.

Objective

To make readers aware of where artificial intelligence has a role in obstetrics and gynecology.

Material and methods

A comprehensive review of the literature was undertaken to compile a list of instances where artificial intelligence was applied to obstetrics and gynecology.

Conclusion

Artificial intelligence should be utilized to benefit patient care and assist the physician in providing data for decision making.

Electrohysterography in the diagnosis of preterm birth: a review

Preterm birth (PTB) is one of the most common and serious complications in pregnancy. About 15 million preterm neonates are born every year, with ratios of 10-15% of total births. In industrialized countries, preterm delivery is responsible for 70% of mortality and 75% of morbidity in the neonatal period. Diagnostic means for its timely risk assessment are lacking and the underlying physiological mechanisms are unclear. Surface recording of the uterine myoelectrical activity (electrohysterogram, EHG) has emerged as a better uterine dynamics monitoring technique than traditional surface pressure recordings and provides information on the condition of uterine muscle in different obstetrical scenarios with emphasis on predicting preterm deliveries.

OBJECTIVE

A comprehensive review of the literature was performed on studies related to the use of the electrohysterogram in the PTB context.

APPROACH

This review presents and discusses the results according to the different types of parameter (temporal and spectral, non-linear and bivariate) used for EHG characterization.

MAIN RESULTS

Electrohysterogram analysis reveals that the uterine electrophysiological changes that precede spontaneous preterm labor are associated with contractions of more intensity, higher frequency content, faster and more organized propagated activity and stronger coupling of different uterine areas. Temporal, spectral, non-linear and bivariate EHG analyses therefore provide useful and complementary information. Classificatory techniques of different types and varying complexity have been developed to diagnose PTB. The information derived from these different types of EHG parameters, either individually or in combination, is able to provide more accurate predictions of PTB than current clinical methods. However, in order to extend EHG to clinical applications, the recording set-up should be simplified, be less intrusive and more robust-and signal analysis should be automated without requiring much supervision and yield physiologically interpretable results.

SIGNIFICANCE

This review provides a general background to PTB and describes how EHG can be used to better understand its underlying physiological mechanisms and improve its prediction. The findings will help future research workers to decide the most appropriate EHG features to be used in their analyses and facilitate future clinical EHG applications in order to improve PTB prediction.

Identification of preterm birth based on RQA analysis of electrohysterograms

BACKGROUND AND OBJECTIVE

Common methods for data analysis are mainly based on linear concepts, but in recent years nonlinear dynamics methods have been introduced. It is a well-known fact that In typical biological systems lack of stationarity and rather sudden changes of state are the properties distinguishing them from each other. There is an urgent need to better understand the mechanical activity of the myometrium (its contractility) to find a solution for preterm delivery problem, the largest cause of neonatal deaths and morbidity. The electrohysterographic signal (EHG) is a good non-linear, bioelectrical indicator for the detection and identification of term and preterm birth.

METHODS

The material of the study consists of EHG signals, obtained from 20 patients between the 24th and the 28th week of pregnancy with threatened preterm labor. The women were divided into two groups: those delivering after more than 7 days - group A (n = 10) and women delivering within 7 days - group B (n = 10). In this paper, an analysis of bioelectrical signals was performed by recurrence quantification analysis (RQA) and principal component analysis (PCA) to distinguish particular patterns for term and preterm birth. To date, these methods have not been used for the evaluation of bioelectrical activity in the uterus. To train novel classifiers for the EHG signals Support Vectors Machine classifications (multiclass SVM) was used. Statistical analysis was performed by means of non-parametric Mann-Whitney test.

RESULTS

From among eleven parameters obtained from recurrence quantification analysis, five most appropriate were chosen: Recurrence Rate, Determinism, Laminarity, Entropy and Recurrence Period Density Entropy. Significant increase (p < .001) of Recurrence Rate was found in patients from group B, while increase of parameters, besides Laminarity, was found in patients from group A. The accuracy of classification obtained as a result of the analysis increased to 83,32%.

CONCLUSION

We showed that the respectively selected recurrence quantificators obtained for that time series could be used to classify all those signals to the appropriate group. The proposed analysis could help in detecting preterm labor based on the EHG signal dynamics.

Machine learning ensemble modelling to classify caesarean section and vaginal delivery types using Cardiotocography traces

Human visual inspection of Cardiotocography traces is used to monitor the foetus during labour and avoid neonatal mortality and morbidity. The problem, however, is that visual interpretation of Cardiotocography traces is subject to high inter and intra observer variability. Incorrect decisions, caused by miss-interpretation, can lead to adverse perinatal outcomes and in severe cases death. This study presents a review of human Cardiotocography trace interpretation and argues that machine learning, used as a decision support system by obstetricians and midwives, may provide an objective measure alongside normal practices. This will help to increase predictive capacity and reduce negative outcomes. A robust methodology is presented for feature set engineering using an open database comprising 552 intrapartum recordings. State-of-the-art in signal processing techniques is applied to raw Cardiotocography foetal heart rate traces to extract 13 features. Those with low discriminative capacity are removed using Recursive Feature Elimination. The dataset is imbalanced with significant differences between the prior probabilities of both normal deliveries and those delivered by caesarean section. This issue is addressed by oversampling the training instances using a synthetic minority oversampling technique to provide a balanced class distribution. Several simple, yet powerful, machine-learning algorithms are trained, using the feature set, and their performance is evaluated with real test data. The results are encouraging using an ensemble classifier comprising Fishers Linear Discriminant Analysis, Random Forest and Support Vector Machine classifiers, with 87% (95% Confidence Interval: 86%, 88%) for Sensitivity, 90% (95% CI: 89%, 91%) for Specificity, and 96% (95% CI: 96%, 97%) for the Area Under the Curve, with a 9% (95% CI: 9%, 10%) Mean Square Error.

Classification of caesarean section and normal vaginal deliveries using foetal heart rate signals and advanced machine learning algorithms

BACKGROUND

Visual inspection of cardiotocography traces by obstetricians and midwives is the gold standard for monitoring the wellbeing of the foetus during antenatal care. However, inter- and intra-observer variability is high with only a 30% positive predictive value for the classification of pathological outcomes. This has a significant negative impact on the perinatal foetus and often results in cardio-pulmonary arrest, brain and vital organ damage, cerebral palsy, hearing, visual and cognitive defects and in severe cases, death. This paper shows that using machine learning and foetal heart rate signals provides direct information about the foetal state and helps to filter the subjective opinions of medical practitioners when used as a decision support tool. The primary aim is to provide a proof-of-concept that demonstrates how machine learning can be used to objectively determine when medical intervention, such as caesarean section, is required and help avoid preventable perinatal deaths.

METHODS

This is evidenced using an open dataset that comprises 506 controls (normal virginal deliveries) and 46 cases (caesarean due to pH ≤ 7.20-acidosis, n = 18; pH > 7.20 and pH < 7.25-foetal deterioration, n = 4; or clinical decision without evidence of pathological outcome measures, n = 24). Several machine-learning algorithms are trained, and validated, using binary classifier performance measures.

RESULTS

The findings show that deep learning classification achieves sensitivity = 94%, specificity = 91%, Area under the curve = 99%, F-score = 100%, and mean square error = 1%.

CONCLUSIONS

The results demonstrate that machine learning significantly improves the efficiency for the detection of caesarean section and normal vaginal deliveries using foetal heart rate signals compared with obstetrician and midwife predictions and systems reported in previous studies.

Prediction of labor onset type: Spontaneous vs induced; role of electrohysterography?

BACKGROUND AND OBJECTIVE

Induction of labor (IOL) is a medical procedure used to initiate uterine contractions to achieve delivery. IOL entails medical risks and has a significant impact on both the mother's and newborn's well-being. The assistance provided by an automatic system to help distinguish patients that will achieve labor spontaneously from those that will need late-term IOL would help clinicians and mothers to take an informed decision about prolonging pregnancy. With this aim, we developed and evaluated predictive models using not only traditional obstetrical data but also electrophysiological parameters derived from the electrohysterogram (EHG).

METHODS

EHG recordings were made on singleton term pregnancies. A set of 10 temporal and spectral parameters was calculated to characterize EHG bursts and a further set of 6 common obstetrical parameters was also considered in the predictive models design. Different models were implemented based on single layer Support Vector Machines (SVM) and with aggregation of majority voting of SVM (double layer), to distinguish between the two groups: term spontaneous labor (≤41 weeks of gestation) and IOL late-term labor. The areas under the curve (AUC) of the models were compared.

RESULTS

The obstetrical and EHG parameters of the two groups did not show statistically significant differences. The best results of non-contextualized single input parameter SVM models were achieved by the Bishop Score (AUC= 0.65) and GA at recording time (AUC= 0.68) obstetrical parameters. The EHG parameter median frequency, when contextualized with the two obstetrical parameters improved these results, reaching AUC= 0.76. Multiple input SVM obtained AUC= 0.70 for all EHG parameters. Aggregation of majority voting of SVM models using contextualized EHG parameters achieved the best result AUC= 0.93.

CONCLUSIONS

Measuring the electrophysiological uterine condition by means of electrohysterographic recordings yielded a promising clinical decision support system for distinguishing patients that will spontaneously achieve active labor before the end of full term from those who will require late term IOL. The importance of considering these EHG measurements in the patient's individual context was also shown by combining EHG parameters with obstetrical parameters. Clinicians considering elective labor induction would benefit from this technique.

Improved Prediction of Preterm Delivery Using Empirical Mode Decomposition Analysis of Uterine Electromyography Signals

Preterm delivery increases the risk of infant mortality and morbidity, and therefore developing reliable methods for predicting its likelihood are of great importance. Previous work using uterine electromyography (EMG) recordings has shown that they may provide a promising and objective way for predicting risk of preterm delivery. However, to date attempts at utilizing computational approaches to achieve sufficient predictive confidence, in terms of area under the curve (AUC) values, have not achieved the high discrimination accuracy that a clinical application requires. In our study, we propose a new analytical approach for assessing the risk of preterm delivery using EMG recordings which firstly employs Empirical Mode Decomposition (EMD) to obtain their Intrinsic Mode Functions (IMF). Next, the entropy values of both instantaneous amplitude and instantaneous frequency of the first ten IMF components are computed in order to derive ratios of these two distinct components as features. Discrimination accuracy of this approach compared to those proposed previously was then calculated using six differently representative classifiers. Finally, three different electrode positions were analyzed for their prediction accuracy of preterm delivery in order to establish which uterine EMG recording location was optimal signal data. Overall, our results show a clear improvement in prediction accuracy of preterm delivery risk compared with previous approaches, achieving an impressive maximum AUC value of 0.986 when using signals from an electrode positioned below the navel. In sum, this provides a promising new method for analyzing uterine EMG signals to permit accurate clinical assessment of preterm delivery risk.

Separating sets of term and pre-term uterine EMG records

The analysis of uterine EMG (electrohysterogram-EHG) records may help solve the problem of predicting pre-term labor. We investigated the adaptive autoregressive (AAR) method to estimate the EHG signal spectrograms and sample entropy, to separate and classify sets of term and pre-term delivery records, using the Term-Preterm EHG Database. The database contains four sets of records divided according to the time of delivery (term or pre-term: ⩾37 or < 37 weeks of gestation, respectively) and according to the time of recording (early or later: before or after the 26th week of gestation, respectively). Using the AAR method the term and pre-term delivery records recorded early can be separated (p = 0.002), as well as all term and pre-term delivery records (p < 0.001). Using the sample entropy, the results showed that all term and pre-term delivery records can be separated (p = 0.022). The spectra of the signals for term delivery records have the tendency of moving to lower frequencies as the time of pregnancy increases. We investigated a few classifiers to classify records between term and pre-term delivery sets. Using median frequency measurements and additional clinical information with the synthetic minority over-sampling technique, the quadratic discriminant analysis classifier achieved a 97% classification accuracy for the records recorded early, and 86% for all records regardless of the time of recording; while for the sample entropy measurements, for the same sets of records, using the support vector machine classifier, the classification accuracies were 80% and 87%, respectively.