Automated 3D ultrasound measurement of the angle of progression in labor

Investigating the relationship between right ventricular size and function with pre-eclampsia: A two-group cross-sectional study

Background and Aims

Pre-eclampsia is a multisystem disorder characterized by symptoms of high blood pressure and proteinuria during pregnancy. It is associated with many complications and maternal and fetal mortality. This disorder may be associated with many cardiovascular complications and affect the function of the heart. Therefore, in this study, the structure and function of the right ventricle (RV) in patients with pre-eclampsia have been investigated using echocardiography.

Methods

This cross-sectional study was conducted in Ghaem Hospital of Mashhad. Thirty-two pregnant women, whose gestational age was 20 weeks or more, were considered as the case group after evaluating blood pressure and confirming proteinuria and pre-eclampsia. Thirty-two healthy pregnant women were also included in the study as a control group. The function of the RV was evaluated using two-dimensional transthoracic echocardiography.

Results

Investigating the results of the study shows that in pregnant women with pre-eclampsia, RV fractional area change, and RV strain indices have decreased significantly compared with healthy pregnant women (p < 0.05). Also, the statistical analysis shows that no significant differences were observed in the two groups in terms of echocardiographic indices E, A, É, E/É, É/Á, E/A, pulmonary artery pressure, Tricuspid Annular Plane Systolic Excursion, right ventricular diameter, and left ventricle mass index.

Conclusion

According to the results of the study, it can be generally said that pre-eclampsia may be associated with changes in the function and echocardiographic indices of the RV and may result in cardiac complications.

A framework for computing angle of progression from transperineal ultrasound images for evaluating fetal head descent using a novel double branch network

Background: Accurate assessment of fetal descent by monitoring the fetal head (FH) station remains a clinical challenge in guiding obstetric management. Angle of progression (AoP) has been suggested to be a reliable and reproducible parameter for the assessment of FH descent. Methods: A novel framework, including image segmentation, target fitting and AoP calculation, is proposed for evaluating fetal descent. For image segmentation, this study presents a novel double branch segmentation network (DBSN), which consists of two parts: an encoding part receives image input, and a decoding part composed of deformable convolutional blocks and ordinary convolutional blocks. The decoding part includes the lower and upper branches, and the feature map of the lower branch is used as the input of the upper branch to assist the upper branch in decoding after being constrained by the attention gate (AG). Given an original transperineal ultrasound (TPU) image, areas of the pubic symphysis (PS) and FH are firstly segmented using the proposed DBSN, the ellipse contours of segmented regions are secondly fitted with the least square method, and three endpoints are finally determined for calculating AoP. Results: Our private dataset with 313 transperineal ultrasound (TPU) images was used for model evaluation with 5-fold cross-validation. The proposed method achieves the highest Dice coefficient (93.4%), the smallest Average Surface Distance (6.268 pixels) and the lowest AoP difference (5.993°) by comparing four state-of-the-art methods. Similar results (Dice coefficient: 91.7%, Average Surface Distance: 7.729 pixels: AoP difference: 5.110°) were obtained on a public dataset with >3,700 TPU images for evaluating its generalization performance. Conclusion: The proposed framework may be used for the automatic measurement of AoP with high accuracy and generalization performance. However, its clinical availability needs to be further evaluated.

The correlation between levator ani co-activation and fetal head regression on maternal pushing at term

OBJECTIVE

To assess the correlation between fetal head regression and levator ani muscle (LAM) co-activation under Valsalva maneuver.

STUDY DESIGN

This study was a secondary analysis of a prospective cohort study on the association between the angle of progression (AoP) and labor outcome. We scanned a group of nulliparous women at term before the onset of labor at rest and under maximum Valsalva maneuver. In addition to the previously calculated AoP, in the present study, we measured the anteroposterior diameter of LAM hiatus (APD) on each ultrasound image. LAM co-activation was defined as APD at Valsalva less than that at rest, whereas fetal head regression was defined as AoP at Valsalva less than that at rest. We calculated the correlation between the two phenomena. Finally, we examined various labor outcomes according to the presence, absence, or co-existence of these two phenomena.

RESULTS

We included 469 women. A total of 129 (27.5%) women presented LAM co-activation while 50 (10.7%) showed head regression. Only 15 (3.2%) women showed simultaneous head regression and LAM co-activation. Women with coexisting LAM co-activation and head regression had the narrowest AoP at Valsalva in comparison with other study groups (p < .001). In addition, they had the highest risk of Cesarean delivery (40%) and longest first, second, and active second stage durations, although none of these reached statistical significance.

CONCLUSION

In nulliparous women at term before the onset of labor fetal head regression and LAM co-activation at Valsalva are two distinct phenomena that uncommonly coexist.

Novel artificial intelligence approach for automatic differentiation of fetal occiput anterior and non-occiput anterior positions during labor

OBJECTIVES

To describe a newly developed machine-learning (ML) algorithm for the automatic recognition of fetal head position using transperineal ultrasound (TPU) during the second stage of labor and to describe its performance in differentiating between occiput anterior (OA) and non-OA positions.

METHODS

This was a prospective cohort study including singleton term (> 37 weeks of gestation) pregnancies in the second stage of labor, with a non-anomalous fetus in cephalic presentation. Transabdominal ultrasound was performed to determine whether the fetal head position was OA or non-OA. For each case, one sonographic image of the fetal head was then acquired in an axial plane using TPU and saved for later offline analysis. Using the transabdominal sonographic diagnosis as the gold standard, a ML algorithm based on a pattern-recognition feed-forward neural network was trained on the TPU images to discriminate between OA and non-OA positions. In the training phase, the model tuned its parameters to approximate the training data (i.e. the training dataset) such that it would identify correctly the fetal head position, by exploiting geometric, morphological and intensity-based features of the images. In the testing phase, the algorithm was blinded to the occiput position as determined by transabdominal ultrasound. Using the test dataset, the ability of the ML algorithm to differentiate OA from non-OA fetal positions was assessed in terms of diagnostic accuracy. The F₁ -score and precision-recall area under the curve (PR-AUC) were calculated to assess the algorithm's performance. Cohen's kappa (κ) was calculated to evaluate the agreement between the algorithm and the gold standard.

RESULTS

Over a period of 24 months (February 2018 to January 2020), at 15 maternity hospitals affiliated to the International Study group on Labor ANd Delivery Sonography (ISLANDS), we enrolled into the study 1219 women in the second stage of labor. On the basis of transabdominal ultrasound, they were classified as OA (n = 801 (65.7%)) or non-OA (n = 418 (34.3%)). From the entire cohort (OA and non-OA), approximately 70% (n = 824) of the patients were assigned randomly to the training dataset and the rest (n = 395) were used as the test dataset. The ML-based algorithm correctly classified the fetal occiput position in 90.4% (357/395) of the test dataset, including 224/246 with OA (91.1%) and 133/149 with non-OA (89.3%) fetal head position. Evaluation of the algorithm's performance gave an F₁ -score of 88.7% and a PR-AUC of 85.4%. The algorithm showed a balanced performance in the recognition of both OA and non-OA positions. The robustness of the algorithm was confirmed by high agreement with the gold standard (κ = 0.81; P < 0.0001).

CONCLUSIONS

This newly developed ML-based algorithm for the automatic assessment of fetal head position using TPU can differentiate accurately, in most cases, between OA and non-OA positions in the second stage of labor. This algorithm has the potential to support not only obstetricians but also midwives and accoucheurs in the clinical use of TPU to determine fetal occiput position in the labor ward. © 2021 International Society of Ultrasound in Obstetrics and Gynecology.

The role of the angle of progression in the prediction of the outcome of occiput posterior position in the second stage of labor

BACKGROUND

Occiput posterior position is the most frequent cephalic malposition, and its persistence at delivery is associated with a higher risk of maternal and perinatal morbidity. Diagnosis and management of occiput posterior position remain a clinical challenge. This is partly caused by our inability to predict fetuses who will spontaneously rotate into occiput anterior from those who will have persistent occiput posterior position. The angle of progression, measured with transperineal ultrasound, represents a reliable tool for the evaluation of fetal head station during labor. The relationship between the persistence of occiput posterior position and fetal head station in the second stage of labor has not been previously assessed.

OBJECTIVE

This study aimed to evaluate the role of fetal head station, as measured by the angle of progression, in the prediction of persistent occiput posterior position and the mode of delivery in the second stage of labor.

STUDY DESIGN

We recruited a nonconsecutive series of women with posterior occiput position diagnosed by transabdominal ultrasound in the second stage of labor. For each woman, a transperineal ultrasound was performed to measure the angle of progression at rest. We compared the angle of progression between women who delivered fetuses in occiput anterior position and those with persistent occiput posterior position at delivery. Receiver operating characteristics curves were performed to evaluate the accuracy of the angle of progression in the prediction of persistent occiput posterior position. Finally, we performed a multivariate logistic regression to determine independent predictors of persistent occiput posterior position.

RESULTS

Overall, 63 women were included in the analysis. Among these, 39 women (62%) delivered in occiput anterior position, whereas 24 (38%) delivered in occiput posterior position (persistent occiput posterior position). The angle of progression was significantly narrower in the persistent occiput posterior position group than in women who delivered fetuses in occiput anterior position (118.3°±12.2° vs 127.5°±10.5°; P=.003). The area under the receiver operating characteristics curve was 0.731 (95% confidence interval, 0.594-0.869) with an estimated best cutoff range of 121.5° (sensitivity of 72% and specificity of 67%). On logistic regression analysis, the angle of progression was found to be independently associated with persistence of occiput posterior position (odds ratio, 0.942; 95% confidence interval, 0.889-0.998; P=.04). Finally, women who underwent cesarean delivery had significantly narrower angle of progression than women who had a vaginal delivery (113.5°±8.1 vs 128.0°±10.7; P<.001). The area under the receiver operating characteristics curve for the prediction of cesarean delivery was 0.866 (95% confidence interval, 0.761-0.972). At multivariable logistic regression analysis including the angle of progression, parity, and gestational age at delivery, the angle of progression was found to be the only independent predictor associated with cesarean delivery (odds ratio, 0.849; 95% confidence interval, 0.775-0.0930; P<.001).

CONCLUSION

In fetuses with occiput posterior at the beginning of the second stage of labor, narrower values of the angle of progression are associated with higher rates of persistent occiput posterior position at delivery and a higher risk of cesarean delivery.

Transperineal ultrasound assessment of maternal pelvic floor at term and fetal head engagement

OBJECTIVES

To evaluate the association between pelvic floor dimensions in nulliparous women at term and fetal head engagement, as assessed by transperineal ultrasound.

METHODS

This was a prospective observational study of nulliparous women at term. Before the onset of labor, transperineal ultrasound was used to measure the anteroposterior diameter (APD) of the levator hiatus and the angle of progression (AoP) at rest, on maximum pelvic floor muscle contraction and on maximum Valsalva maneuver (before and after visual feedback). We assessed the correlation between pelvic floor static and dynamic dimensions (levator hiatal APD and levator ani muscle coactivation) and AoP, which is an objective index of fetal head engagement.

RESULTS

In total, 282 women were included in the analysis. Among these, 211 (74.8%) women had a vaginal delivery while 71 (25.2%) had a Cesarean delivery. AoP was narrower in the Cesarean-delivery group at rest, on maximum pelvic floor muscle contraction and on maximum Valsalva, whereas no differences in levator hiatal APD were found between the two groups. We found a negative correlation between levator hiatal APD at rest, on maximum pelvic floor muscle contraction and on maximum Valsalva and the duration of the second stage of labor. There was a positive correlation between AoP and levator hiatal APD on maximum Valsalva maneuver after visual feedback (r = 0.15, P = 0.01). Women with levator ani muscle contraction on Valsalva maneuver (i.e. coactivation), both pre and post visual feedback, had a narrower AoP at rest and on maximum Valsalva. After visual feedback, women with levator ani muscle coactivation had a longer second stage of labor than did those without (80.8 ± 61.4 min vs 62.9 ± 43.4 min (P = 0.04)).

CONCLUSIONS

Smaller pelvic floor dimensions and levator ani muscle coactivation are associated with higher fetal head station and with a longer second stage of labor in nulliparous women at term. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.

Automatic measurement of head-perineum distance during intrapartum ultrasound: description of the technique and preliminary results

OBJECTIVES

To evaluate the accuracy and reliability of a new ultrasound technique for the automatic assessment of the head-perineum distance (HPD) during childbirth.

METHODS

HPD was measured on a total of 40 acquisition sessions in 30 laboring women both automatically by an innovative algorithm and manually by trained sonographers, assumed as gold standard.

RESULTS

A significant correlation was found between manual and automatic measurements (Intra-CC = 0.994). High values of the coefficient of determination (r²=0.98) and low residual errors: RMSE = 2.01 mm (4.9%) were found.

CONCLUSION

The automatic algorithm for the assessment of the HPD represents a reliable technique.

Comparison of different methods of measuring angle of progression in prediction of labor outcome

OBJECTIVES

First, to compare the manual sagittal and parasagittal and automated parasagittal methods of measuring the angle of progression (AoP) by transperineal ultrasound during labor, and, second, to develop models for the prediction of time to delivery and need for Cesarean section (CS) for failure to progress (FTP) in a population of patients undergoing induction of labor.

METHODS

This was a prospective observational study of transperineal ultrasound in a cohort of 512 women with a singleton pregnancy undergoing induction of labor. A random selection of 50 stored images was assessed for inter- and intraobserver reliability of AoP measurements using the manual sagittal and parasagittal and automated parasagittal methods. In cases of vaginal delivery, univariate linear, multiple linear and quantile regression analyses were performed to predict time to delivery. Univariate and multivariate binomial logistic regression analyses were performed to predict CS for FTP in the first stage of labor.

RESULTS

The intraclass correlation coefficient (ICC) for the manual parasagittal method for a single observer was 0.97 (95% CI, 0.95-0.98) and for two observers it was 0.96 (95% CI, 0.93-0.98), indicating good reliability. The ICC for the sagittal method for a single observer was 0.93 (95% CI, 0.88-0.96) and for two observers it was 0.74 (95% CI, 0.58-0.84), indicating moderate reliability for a single observer and poor reliability between two observers. Bland-Altman analysis demonstrated narrower limits of agreement for the manual parasagittal approach than for the sagittal approach for both a single and two observers. The automated parasagittal method failed to capture an image in 19% of cases. The mean difference in AoP measurements between the sagittal and manual parasagittal methods was 11°. In pregnancies resulting in vaginal delivery, 54% of the variation in time to delivery was explained in a model combining parity, epidural and syntocinon use during labor and the sonographic findings of fetal head position and AoP. In the prediction of CS for FTP in the first stage of labor, a model which combined maternal factors with the sonographic measurements of AoP and estimated fetal weight was superior to one utilizing maternal factors alone (area under the receiver-operating-characteristics curve, 0.80 vs 0.76).

CONCLUSIONS

First, the method of measuring AoP with the greatest reliability is the manual parasagittal technique and future research should focus on this technique. Second, over half of the variation in time to vaginal delivery can be explained by a model that combines maternal factors, pregnancy characteristics and ultrasound findings. Third, the ability of AoP to provide clinically useful prediction of CS for FTP in the first stage of labor is limited. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

Reliability of transperineal ultrasound for the assessment of the angle of progression in labor using parasagittal approach versus midsagittal approach

OBJECTIVE

To assess the inter-method agreement between midsagittal (msAoP) and parasagittal (psAoP) measurements of the angle of progression (AoP) during labor. In addition, we aimed to evaluate the correlation between AoP measurements by both midsagittal and parasagittal approaches with the mode of delivery.

METHODS

We recruited a nonconsecutive series of women in active labor with a singleton uncomplicated term pregnancy with fetuses in vertex presentation. Women underwent transperineal ultrasound in the absence of uterine contractions or maternal pushing to measure both msAoP and psAoP. The inter-method agreement between the two acquisitions was then assessed. Lastly, both measurements were compared between women who had a vaginal delivery versus those who underwent cesarean section (CS).

RESULTS

Overall, 151 women were included in the study. We found an excellent agreement between msAoP and psAoP (ICC 0.935; 95% CI 0.912-0.953, p < .001). On the other hand, psAoP overestimated the measurements in comparison with msAoP (101.2 ± 15.6 versus 98.2 ± 16.0, p < .001). There was a significant correlation between both methods of AoP assessment and duration of the active second stage of labor and AoP measured by either method was significantly wider in patients who delivered vaginally compared to those who had a CS.

CONCLUSIONS

Our data showed a significant difference in the measured angle between the psAoP and the originally described msAoP. The automated measurements of AoP that have been introduced are designed using the parasagittal visualization of the more echogenic pubic arch, rather than the hypoechogenic pubic symphysis. We think that in the light of our data, care should be taken before applying data from midsagittal measurement in centers using the parasagittal automated approach.

Knowledge about Neonatal Screening among Postpartum Women and Complexity Level of Birthing Facilities

BACKGROUND

To ascertain the degree of knowledge of postpartum women about important aspects related to the neonatal screening process and whether differences of opinion exist between those who deliver in low-complexity versus high-complexity health facilities (low-risk versus high-risk pregnancies, respectively).

METHODS

This was a prospective, cross-sectional, questionnaire-based study. The sample consisted of postpartum women recruited from 2013 to 2015 at public maternity hospitals in the city of Niterói, Brazil. Participants were divided into two groups and completed a questionnaire consisting of Likert-scored items. Continuous variables were analyzed with the Mann-Whitney test, and categorical variables, with Fisher's test. A confirmatory factor analysis of participants' answers was performed.

RESULTS

Of 188 women enrolled, 54 (28.7%) had incomplete elementary education; 119 (62.2%) had attended more than six antenatal care visits. The mean age was 25.57 years. Nearly all women (n = 179, 95.2%) were roomed-in with their infants. Knowledge of neonatal screening was very similar in the high-complexity and low-complexity groups. Divergences were limited to items regarding the risks of neonatal screening.

CONCLUSIONS

The degree of knowledge among postpartum women was similar among high- and low-complexity facilities. Those who attended high-complexity facilities had longer hospital stays and greater adherence to ethical issues regarding neonatal screening.

Ultrasound in labor: impact of a theoretical and practical course on caregiver's perspective and accuracy

Objective: The aim of the present study was two-fold; first is to assess the impact of a theoretical and practical course on the caregiver's perspective on the use of ultrasound in labor. Second is to evaluate the impact of the course on the measurement of the angle of progression.Materials and methods: Obstetricians participating in the course responded twice to a questionnaire addressing the caregiver's perspective on the use of ultrasound in labor, before and at the end of the course. In addition, all participants measured the angle of progression on a digital simulation software once following the theoretical part and a second time after the conclusion of both the theoretical and practical sessions. All answers to the questionnaire were compared before versus after the course. The angle of progression assessments after the theoretical part and those after the course were compared with those performed by an expert operator.Results: Overall, data from 51 participants were analyzed. In comparison with the pre-course questionnaire, more participants after the course believed ultrasound can be a useful instrument for fetal occiput presentation diagnosis (45; 88.2% versus 26; 51%, p < .001) and for fetal head progression evaluation (37; 72.5% versus 19; 37.3%; p < .001). Although before the course, most participants (32; 62.7%) thought that the transperineal parameters were too complex to be applied, only 19 (37.3%) still considered the complexity of the ultrasound indices, an obstacle against their application in clinical practice after the course (p = .002). Regarding the angle of progression, precourse measurements showed a systematic overestimation in comparison with the gold standard. Furthermore, the coefficient of variation, more than halved following the practical simulation reflecting at least a doubling of the angle of progression (AoP) measurement precision (14.2 versus 5.5%).Conclusions: A theoretical and practical course conducted by expert operators may positively change obstetricians' perspective on the use of ultrasound in labor. Moreover, a practical session seems to improve the participants' accuracy and precision in the assessment of fetal head descent by transperineal ultrasound.