End-to-end learning with interpretation on electrohysterography data to predict preterm birth.

Prediction of preterm birth is a difficult task for clinicians. By examining an electrohysterogram, electrical activity of the uterus that can lead to preterm birth can be detected. Since signals associated with uterine activity are difficult to interpret for clinicians without a background in signal processing, machine learning may be a viable solution. We are the first to employ Deep Learning models, a long-short term memory and temporal convolutional network model, on electrohysterography data using the Term-Preterm Electrohysterogram database. We show that end-to-end learning achieves an AUC score of 0.58, which is comparable to machine learning models that use handcrafted features. Moreover, we evaluate the effect of adding clinical data to the model and conclude that adding the available clinical data to electrohysterography data does not result in a gain in performance. Also, we propose an interpretability framework for time series classification that is well-suited to use in case of limited data, as opposed to existing methods that require large amounts of data. Clinicians with extensive work experience as gynaecologist used our framework to provide insights on how to link our results to clinical practice and stress that in order to decrease the number of false positives, a dataset with patients at high risk of preterm birth should be collected. All code is made publicly available.

Risk of preterm birth after prior term cesarean.

OBJECTIVE: To determine the risk of overall preterm birth (PTB) and spontaneous PTB in a pregnancy after a caesarean section (CS) at term. DESIGN: Longitudinal linked national cohort study. SETTING: The Dutch Perinatal Registry (1999-2009). POPULATION: 268 495 women with two subsequent singleton pregnancies were identified. METHODS: A cohort study based on linked registered data from two subsequent pregnancies in the Netherlands. MAIN OUTCOME MEASURES: The incidence of overall PTB and spontaneous PTB with subgroup analysis on gestational age at first delivery and type of CS (planned or unplanned). RESULTS: Of 268 495 women with a singleton first pregnancy who delivered at term, 15.76% (n = 42 328) had a CS. The incidence of PTB in the second pregnancy was 2.79% (n = 1182) in women with a previous CS versus 2.46% (n = 5570) in women with a previous vaginal delivery (adjusted odds ratio [aOR] 1.14, 95% confidence interval [CI] 1.07-1.21). This increased risk is mainly driven by an increased risk of spontaneous PTB after previous CS at term (aOR 1.50, 95% CI 1.38-1.70). Analysis for type of CS compared with vaginal delivery showed an aOR on spontaneous PTB of 1.86 (95% CI 1.58-2.18) for planned CS and an aOR of 1.40 (95% CI 1.24-1.58) for unplanned CS. CONCLUSIONS: CS at term is associated with a marginally increased risk of spontaneous PTB in a subsequent pregnancy. TWEETABLE ABSTRACT: Caesarean section at term is associated with a marginally increased risk of spontaneous PTB in a subsequent pregnancy.

Effect of interpregnancy interval on the success rate of trial of labor after cesarean.

OBJECTIVE: The objective of this study is to investigate the association between interpregnancy interval and success of vaginal birth after cesarean. STUDY DESIGN: Retrospective 10-year cohort study of pregnant women with one prior cesarean, who opted for trial of labor (n=36 653). Interpregnancy interval is the time between cesarean and next conception. Vaginal birth success rates were compared between six interval groups. Analysis was performed pooled as well as stratified for induction of labor. Adjusted odds ratios were calculated. RESULTS: Success rate in the reference group (12 to 24 months) was 72%. Success rates were similar among those with an interval of less than 24 months. Intervals of 24 months or more showed a decrease in success rate; 70% in 24- to 35-month intervals (adjusted odds ratio 0.92 (0.87 to 0.98)), 67% in 36- to 59-month intervals (adjusted odds ratio 0.87 (0.81 to 0.94)) and 62% in intervals of more than 60 months (adjusted odds ratio 0.77 (0.67 to 0.88)). CONCLUSION: An interpregnancy interval of <24 months is not associated with a decreased success of vaginal birth after cesarean. Success rates decrease when interval increases.

Trial of labor after cesarean: attempted operative vaginal delivery versus emergency repeat cesarean, a prospective national cohort study.

OBJECTIVE: To compare neonatal and maternal outcomes of attempted operative vaginal delivery with emergency repeat cesarean in trial of labor after cesarean. STUDY DESIGN: Prospective 8-year cohort analysis using the Netherlands Perinatal Registry, including women with one prior cesarean giving birth through operative vaginal delivery or emergency repeat cesarean (n=12860). A multivariate analysis was performed. Odds ratios (OR) and adjusted odds ratios (aOR) were calculated. RESULTS: Attempted operative vaginal delivery increases the risk on neonatal birth trauma (aOR 15.0 (5.94 to 38.0)) and postpartum hemorrhage (aOR 2.59 (2.17 to 3.09)), and lowers the risk of wet lung syndrome (aOR 0.53 (0.35 to 0.80)) and neonatal convulsions (aOR 0.47 (0.24 to 0.91)). CONCLUSION: We found a highly increased risk of neonatal birth trauma and a moderately increased risk of postpartum hemorrhage but slightly lower risks of wet lung syndrome and neonatal convulsions after attempted operative vaginal delivery compared with emergency repeat cesarean.