Quantitative analysis of cervical texture by ultrasound in mid-pregnancy and association with spontaneous preterm birth

Assessment of novel sonographic and biochemical tools for spontaneous preterm birth prediction in asymptomatic twin pregnancies

INTRODUCTION

Prematurity is a major global health issue. Twin pregnancies are a group at especially high risk of preterm birth. Sonographic mid-trimester cervical length has limited accuracy in predicting preterm birth. This study aimed to evaluate the association between mid-trimester sonographic markers of early cervical remodeling and cervical inflammatory biomarkers and fetal fibronectin, alone or in combination, as predictors of preterm birth before 34+0 weeks in asymptomatic twin pregnancies.

MATERIAL AND METHODS

Prospective cohort study, including uncomplicated dichorionic or monochorionic-diamniotic twin pregnancies, recruited and assessed between 18+0 and 24+6 weeks, from a single tertiary referral center between 2020 and 2023. At inclusion, transvaginal ultrasound was performed to assess the following sonographic markers (cervical length, uterocervical angle, cervical consistency index, cervical texture) and an endocervical sample was obtained prior to ultrasound to quantify the following cervical inflammatory biomarkers (tumor necrosis factor alpha, interleukins 1b, 6, 8, 18, matrix metalloproteinase-8 and 9) and fetal fibronectin. The diagnostic performance of those sonographic and biochemical markers independently associated with spontaneous preterm birth before 34 weeks was analyzed by receiver operating characteristic curves and assessed through sensitivity and specificity analysis for several cutoffs.

RESULTS

Of the 172 women included, cervical length was shorter (36 mm vs. 40 mm; p = 0.025) and uterocervical angle was wider (137° vs. 120°; p = 0.004) in the preterm group. Cervical consistency index, cervical texture score, cervical inflammatory biomarkers, and fetal fibronectin were similar among the study groups. The area under the curve to predict spontaneous preterm birth before 34+0 weeks was 0.722 (95% CI 0.577 to 0.866) for cervical length, 0.789 (95% CI 0.683 to 0.895) for uterocervical angle, and 0.852 (95% CI 0.752 to 0.952) for a combination of both. Based on the receiver operating characteristics curve cutoff, sensitivity and specificity for cervical length ≤37 mm was 55.6% and 66.3%, for an uterocervical angle ≥135° was 77.8% and 76.1%, and for both criteria present 44.4% and 93.3%, respectively.

CONCLUSIONS

This finding of this study suggests that the combination of cervical length and uterocervical angle in mid-trimester sonographic assessment may improve the prediction of preterm birth before 34 weeks in asymptomatic and uncomplicated twin pregnancies.

Techniques for detecting cervical remodeling as a predictor for spontaneous preterm birth: current evidence and future research avenues in patients with multiple pregnancies

BACKGROUND

Spontaneous preterm birth occurs more frequently in multiple pregnancies. This syndrome has multiple triggers that result in a unified downstream pathway of cervical remodeling, uterine activity, and progressive cervical dilatation. Whilst the triggers for labor in multiple pregnancy may be different from singletons, the downstream changes will be the same. Identifying patients at risk of preterm birth is a priority as interventions to delay delivery and optimize the fetus can be initiated. Methods for screening for risk of preterm birth which focus on the detection of cervical remodeling may therefore have potential in this population.

METHODS

This review explores the evidence for the predictive utility for preterm birth of several published techniques that assess the physical, biomechanical, and optical properties of the cervix, with a focus on those which have been studied in multiple pregnancies and highlighting targets for future research in this population.

RESULTS

Fifteen techniques are discussed which assess the physical, biomechanical, and optical properties of the cervix in pregnancy. Of these, only three techniques that evaluated the predictive accuracy of a technique in patients with multiple pregnancies were identified: uterocervical angle, cervical consistency index, and cervical elastography. Of these, measurement of the uterocervical angle has the strongest evidence. Several techniques have shown predictive potential in singleton pregnancies, but have not yet been studied in multiple pregnancies, which would be a logical expansion of research.

CONCLUSION

Research on techniques with predictive utility for PTB in patients with multiple pregnancies is limited but should be a research priority. Overall, the theory supports the investigation of cervical remodeling as a predictor of PTB, and there are numerous techniques in development that may have potential in this field.

Risk Scoring Systems for Preterm Birth and Their Performance: A Systematic Review

Introduction: Nowadays, the risk stratification of preterm birth (PTB) and its prediction remain a challenge. Many risk factors associated with PTB have been identified, and risk scoring systems (RSSs) have been developed to face this challenge. The objectives of this systematic review were to identify RSSs for PTB, the variables they consist of, and their performance. Materials and methods: Two databases were searched, and two authors independently performed the screening and eligibility phases. Records studying an RSS, based on specified variables, with an evaluation of the predictive value for PTB, were considered eligible. Reference lists of eligible studies and review articles were also searched. Data from the included studies were extracted. Results: A total of 56 studies were included in this review. The most frequently incorporated variables in the RSS included in this review were maternal age, weight, history of smoking, history of previous PTB, and cervical length. The performance measures varied widely among the studies, with sensitivity ranging between 4.2% and 92.0% and area under the curve (AUC) between 0.59 and 0.95. Conclusions: Despite the recent technological and scientifical evolution with a better understanding of variables related to PTB and the definition of new ultrasonographic parameters and biomarkers associated with PTB, the RSS's ability to predict PTB remains poor in most situations, thus compromising the integration of a single RSS in clinical practice. The development of new RSSs, the identification of new variables associated with PTB, and the elaboration of a large reference dataset might be a step forward to tackle the problem of PTB.

Machine learning and disease prediction in obstetrics

Machine learning technologies and translation of artificial intelligence tools to enhance the patient experience are changing obstetric and maternity care. An increasing number of predictive tools have been developed with data sourced from electronic health records, diagnostic imaging and digital devices. In this review, we explore the latest tools of machine learning, the algorithms to establish prediction models and the challenges to assess fetal well-being, predict and diagnose obstetric diseases such as gestational diabetes, pre-eclampsia, preterm birth and fetal growth restriction. We discuss the rapid growth of machine learning approaches and intelligent tools for automated diagnostic imaging of fetal anomalies and to asses fetoplacental and cervix function using ultrasound and magnetic resonance imaging. In prenatal diagnosis, we discuss intelligent tools for magnetic resonance imaging sequencing of the fetus, placenta and cervix to reduce the risk of preterm birth. Finally, the use of machine learning to improve safety standards in intrapartum care and early detection of complications will be discussed. The demand for technologies to enhance diagnosis and treatment in obstetrics and maternity should improve frameworks for patient safety and enhance clinical practice.

Artificial Intelligence in Obstetric Ultrasound: An Update and Future Applications

Artificial intelligence (AI) can support clinical decisions and provide quality assurance for images. Although ultrasonography is commonly used in the field of obstetrics and gynecology, the use of AI is still in a stage of infancy. Nevertheless, in repetitive ultrasound examinations, such as those involving automatic positioning and identification of fetal structures, prediction of gestational age (GA), and real-time image quality assurance, AI has great potential. To realize its application, it is necessary to promote interdisciplinary communication between AI developers and sonographers. In this review, we outlined the benefits of AI technology in obstetric ultrasound diagnosis by optimizing image acquisition, quantification, segmentation, and location identification, which can be helpful for obstetric ultrasound diagnosis in different periods of pregnancy.

Prediction and Prevention of Spontaneous Preterm Birth: ACOG Practice Bulletin, Number 234

Preterm birth is among the most complex and important challenges in obstetrics. Despite decades of research and clinical advancement, approximately 1 in 10 newborns in the United States is born prematurely. These newborns account for approximately three-quarters of perinatal mortality and more than one half of long-term neonatal morbidity, at significant social and economic cost (1-3). Because preterm birth is the common endpoint for multiple pathophysiologic processes, detailed classification schemes for preterm birth phenotype and etiology have been proposed (4, 5). In general, approximately one half of preterm births follow spontaneous preterm labor, about a quarter follow preterm prelabor rupture of membranes (PPROM), and the remaining quarter of preterm births are intentional, medically indicated by maternal or fetal complications. There are pronounced racial disparities in the preterm birth rate in the United States. The purpose of this document is to describe the risk factors, screening methods, and treatments for preventing spontaneous preterm birth, and to review the evidence supporting their roles in clinical practice. This Practice Bulletin has been updated to include information on increasing rates of preterm birth in the United States, disparities in preterm birth rates, and approaches to screening and prevention strategies for patients at risk for spontaneous preterm birth.

Mid-trimester prediction of spontaneous preterm birth with automated cervical quantitative ultrasound texture analysis and cervical length: a prospective study

The objective of this study was to evaluate a novel automated test based on ultrasound cervical texture analysis to predict spontaneous Preterm Birth (sPTB) alone and in combination with Cervical Length (CL). General population singleton pregnancies between 18 + 0 and 24 + 6 weeks’ gestation were assessed prospectively at two centers. Cervical ultrasound images were evaluated and the occurrence of sPTB before weeks 37 + 0 and 34 + 0 were recorded. CL was measured on-site. The automated texture analysis test was applied offline to all images. Their performance to predict the occurrence of sPTB before 37 + 0 and 34 + 0 weeks was evaluated separately and in combination on 633 recruited patients. AUC for sPTB prediction before weeks 37 and 34 respectively were as follows: 55.5% and 65.3% for CL, 63.4% and 66.3% for texture analysis, 67.5% and 76.7% when combined. The new test improved detection rates of CL at similar low FPR. Combining the two increased detection rate compared to CL alone from 13.0 to 30.4% for sPTB < 37 and from 14.3 to 42.9% sPTB < 34. Texture analysis of cervical ultrasound improved sPTB detection rate compared to cervical length for similar FPR, and the two combined together increased significantly prediction performance. This results should be confirmed in larger cohorts.

Magnetic resonance diffusion tensor imaging of cervical microstructure in normal early and late pregnancy in vivo

BACKGROUND

Cervical remodeling is an important aspect of birth timing. Before cervical ripening, the collagen fibers are arranged in a closely interweaved network, but during ripening, the fibers become disorganized and the cervix becomes more hydrated. To quantitatively measure cervical remodeling, we need a noninvasive method to monitor changes in cervical collagen fiber organization and hydration in vivo.

OBJECTIVE

To use diffusion tensor imaging to image and quantify the spatial and temporal differences in cervical microstructure between normal early and late pregnancies.

STUDY DESIGN

After institutional review board approval and consent, a group of healthy women in early pregnancy (22 patients at 12-14 weeks' gestation) and a group in late pregnancy (27 patients at 36-38 weeks' gestation) underwent magnetic resonance imaging on a Siemens MAGNETOM Vida 3 Tesla unit. Diffusion tensor imaging of the cervix in the axial plane was performed with a two-dimensional single-shot echo planar imaging diffusion-weighted sequence. In early and late pregnancy groups, the differences of the diffusion tensor imaging measures were compared between the subglandular zone and the outer stroma regions of the cervix. In addition, the diffusion tensor imaging measures were compared between the early and late pregnancy groups. Finally, for the late pregnancy group, the diffusion tensor imaging measures were compared between the primipara and multipara groups.

RESULTS

Diffusion tensor imaging measures of microstructure significantly differed between the subglandular zone and outer stroma regions of the cervix in both early and late pregnancies. In the subglandular zone, fractional anisotropy was lower in the late pregnancy group than in the early pregnancy group (0.37 [0.34-0.42] vs 0.50 [0.43-0.58]; P<.0005), suggesting increased collagen fiber disorganization in this zone. In addition, mean diffusivity was higher in the late pregnancy group than in the early pregnancy group (1.84 [1.73-2.02] mm²/sec×10^-3 vs 1.56 [1.42-1.69] mm²/sec×10^-3; P=.001), suggesting increased hydration in the subglandular zone. In the outer stroma, neither fractional anisotropy (0.44 [0.40-0.50] vs 0.41 [0.37-0.43]; P=.095) nor mean diffusivity (2.09 [1.92-2.25] mm²/sec×10^-3 vs 2.12 [2.04-2.24] mm²/sec×10^-3; P=.269) significantly differed between early pregnancy and late pregnancy, suggesting insignificant temporal microstructural changes in this cervical zone. Diffusion tensor imaging measures did not significantly differ between cervixes from primiparous and multiparous women in late pregnancy.

CONCLUSION

This in vivo study demonstrates that diffusion tensor imaging can noninvasively quantify the microstructural differences in collagen fiber organization and hydration in cervical subregions between early pregnancy and late pregnancy.

Uterine Cervical Change at Term Examined Using Ultrasound Elastography: A Longitudinal Study

The aim of the study was to investigate if there are changes in elastographic parameters in the cervix at term around the time of delivery and if there are differences in the parameters between women with spontaneous labor and those without labor (labor induction). Nulliparous women at 36 weeks of gestation eligible for vaginal delivery were enrolled. Cervical elastography was performed and cervical length were measured using the E-Cervix^TM system (WS80A Ultrasound System, Samsung Medison, Seoul, Korea) at each weekly antenatal visit until admission for spontaneous labor or labor induction. E-Cervix parameters of interest included elasticity contrast index (ECI), internal os strain mean level (IOS), external os strain mean level (EOS), IOS/EOS strain mean ratio, strain mean level, and hardness ratio. Regression analysis was performed using days from elastographic measurement at each visit to admission for delivery and the presence or absence of labor against cervical length, and each E-Cervix parameter fitted to a linear model for longitudinal data measured repeatedly. A total of 96 women were included in the analysis, (spontaneous labor, n = 39; labor induction, n = 57). Baseline characteristics were not different between the two groups except for cesarean delivery rate. Cervical length decreased with advancing gestation and was different between the two groups. Most elastographic parameters including ECI, IOS, EOS, strain mean, and hardness ratio were significantly different between the two groups. In addition, ECI, IOS, and strain mean values significantly increased with advancing gestation. Our longitudinal study using ultrasound elastography indicated that E-cervix parameters tended to change linearly at term near the time of admission for delivery and that there were differences in E-Cervix parameters according to the presence or absence of labor.

A new quantification system for assessing the degree of acute cervical insufficiency based on physical and sonographic examination

OBJECTIVES

Acute cervical insufficiency accounts for 10-25 % of all mid-trimester pregnancy losses. However, the definition and description for the degree of acute cervical insufficiency were obscure and different among the many studies. The aim of this study was to suggest a new 4-digit quantification system and to evaluate the outcome according to the new system in women with acute cervical insufficiency.

STUDY DESIGN

A retrospective cohort study was conducted in patients with acute cervical insufficiency who underwent physical examination indicated cervical cerclage. Acute cervical insufficiency was defined as painless external os dilation with prolapsed and/or visible membranes on speculum examination. The status of fetal membranes was described using two values: 1) size of the prolapsed membrane (P, measured using ultrasound); and 2) size of visible fetal membranes (M, evaluated by speculum examination). The status of cervix was described using two values: 1) dilatation of the narrowest os (O, measured by ultrasound); and 2) functional cervical length (C, measured by ultrasound). The patients were divided into 3 groups as follows: Stage I, patients with visible fetal membranes (M > 0) but with a remaining functional cervix (C > 0) (N = 7); stage II, those with visible fetal membranes (M > 0) and a functional cervical length of 0 but without prolapsed membranes (P = 0) (N = 33); and stage III, those with prolapsed membranes (P > 0) (N = 40).

RESULTS

1) Patients who delivered before 34 weeks of gestation had a significantly lower median gestational age at cerclage operation and a significantly higher median P, M, C values than those who delivered at or beyond 34 weeks of gestation(P < 0.01 for gestational age at operation, P and M values; P < 0.05 for C value); 2) The higher the stage based on PMOC quantification system, the greater the risk of preterm delivery before <24, <32, <34, and <37 weeks of gestation (P < 0.001 for all); 3) The neonatal survival rate was 100 % (7/7) for stage I, 93.9 % (31/33) for stage II, and 60 % (24/40) for stage III.

CONCLUSION

The PMOC system was a simple method to describe the individualized conditions and to predict the risk of preterm births in all spectrums of acute cervical insufficiency.

Evaluation of deep convolutional neural networks for automatic classification of common maternal fetal ultrasound planes

The goal of this study was to evaluate the maturity of current Deep Learning classification techniques for their application in a real maternal-fetal clinical environment. A large dataset of routinely acquired maternal-fetal screening ultrasound images (which will be made publicly available) was collected from two different hospitals by several operators and ultrasound machines. All images were manually labeled by an expert maternal fetal clinician. Images were divided into 6 classes: four of the most widely used fetal anatomical planes (Abdomen, Brain, Femur and Thorax), the mother’s cervix (widely used for prematurity screening) and a general category to include any other less common image plane. Fetal brain images were further categorized into the 3 most common fetal brain planes (Trans-thalamic, Trans-cerebellum, Trans-ventricular) to judge fine grain categorization performance. The final dataset is comprised of over 12,400 images from 1,792 patients, making it the largest ultrasound dataset to date. We then evaluated a wide variety of state-of-the-art deep Convolutional Neural Networks on this dataset and analyzed results in depth, comparing the computational models to research technicians, which are the ones currently performing the task daily. Results indicate for the first time that computational models have similar performance compared to humans when classifying common planes in human fetal examination. However, the dataset leaves the door open on future research to further improve results, especially on fine-grained plane categorization.

The Predictive Value of the Cervical Consistency Index to Predict Spontaneous Preterm Birth in Asymptomatic Twin Pregnancies at the Second-Trimester Ultrasound Scan: A Prospective Cohort Study

Novel transvaginal ultrasound (TVU) markers have been proposed to improve spontaneous preterm birth (sPTB) prediction. Preliminary results of the cervical consistency index (CCI), uterocervical angle (UCA), and cervical texture (CTx) have been promising in singletons. However, in twin pregnancies, the results have been inconsistent. In this prospective cohort study of asymptomatic twin pregnancies assessed between 18⁺⁰–22⁺⁰ weeks, we evaluated TVU derived cervical length (CL), CCI, UCA, and the CTx to predict sPTB < 34⁺⁰ weeks. All iatrogenic PTB were excluded. In the final cohort of 63 pregnancies, the sPTB rate < 34⁺⁰ was 16.3%. The CCI, UCA, and CTx, including the CL was significantly different in the sPTB < 34⁺⁰ weeks group. The best area under the receiver operating characteristic curve (AUC) for sPTB < 34⁺⁰ weeks was achieved by the CCI 0.82 (95%CI, 0.72–0.93), followed by the UCA with AUC 0.72 (95%CI, 0.57–0.87). A logistic regression model incorporating parity, chorionicity, CCI, and UCA resulted in an AUC of 0.91 with a sensitivity of 55.3% and specificity of 88.1% for predicting sPTB < 34⁺⁰. The CCI performed better than other TVU markers to predict sPTB < 34⁺⁰ in twin gestations, and the best diagnostic accuracy was achieved by a combination of parity, chorionicity, CCI, and UCA.

The role of noninvasive diagnostic imaging in monitoring pregnancy and detecting patients at risk for preterm birth: a review of quantitative approaches

Preterm birth (PTB) is the leading cause of neonatal morbidity and mortality worldwide. The ability to predict patients at risk for preterm birth remains a major health challenge. The currently available clinical diagnostics such as cervical length and fetal fibronectin may detect only up to 30% of patients who eventually experience a spontaneous preterm birth. This paper reviews ongoing efforts to improve the ability to conduct a risk assessment for preterm birth. In particular, this work focuses on quantitative methods of imaging using ultrasound-based techniques, magnetic resonance imaging, and optical imaging modalities. While ultrasound imaging is the major modality for preterm birth risk assessment, a summary of efforts to adopt other imaging modalities is also discussed to identify the technical and diagnostic limits associated with adopting them in clinical settings. We conclude the review by proposing a new approach using combined photoacoustic, ultrasound, and elastography as a potential means to better assess cervical tissue remodeling, and thus improve the detection of patients at-risk of PTB.

Quantitative Ultrasound Analysis of Proximal and Distal Cervical Tissue Echogenicity in Premature Cervical Remodeling

OBJECTIVES

To determine whether a novel, noninvasive quantitative ultrasound (US) technique can detect differences in proximal and distal cervical tissue echogenicity in women with and without a shortened cervical length (CL).

METHODS

We conducted a retrospective case-control study of singleton pregnancies at 16 to 26 weeks' gestation in which a transvaginal US examination was performed to measure CL from 2013 to 2015. Initial CLs in cases and controls were less than 2.5 cm and 2.5 cm or greater, respectively. For each US image, a region of interest was selected in the proximal and distal cervical stroma, in both the anterior and posterior cervical lips. The Floyd-Steinberg dithering algorithm transformed grayscale pixels in each region of interest into a binary map. A histogram tabulated the number of black and white pixels, allowing determination of the percent echogenicity. The difference in the percent echogenicity was calculated by subtracting the distal cervical echogenicity (average of anterior and posterior lips) from the proximal cervical echogenicity (average of anterior and posterior lips).

RESULTS

Ultrasound images from 177 women were analyzed. There was a difference in the percent echogenicity (P < .0001) when comparing women with a short cervix (mean ± SD, 9.8 ± 10.1; n = 102) to women with a normal CL (17.2 ± 9.5; n = 75). Differences were attributable to changes in proximal (P < .008) rather than distal cervical echogenicity. Regardless of CL, the proximal cervix was more echogenic than the distal cervix.

CONCLUSIONS

A quantitative US analysis of cervical tissue can detect differences in echogenicity between the proximal and distal cervix in the second trimester. Proximal cervical echogenicity is lower with CL of less than 2.5 cm compared to a normal CL.

Clinical and Sonographic Model to Predict Cesarean Delivery after Induction of Labor at Term

OBJECTIVE

To develop a model combining clinical and sonographic features to predict the risk of cesarean delivery after the induction of labor (IOL).

METHODS

We designed a prospective observational study involving women admitted for IOL. The main outcome was defined as cesarean delivery due to failed IOL or arrest of labor. Several clinical and ultrasonographic variables were collected. Seventy percent of the sample was used to build the predictive model, using stepwise logistic regression, while the remaining sample was used for validation. The final model was estimated and calibrated using all participants.

RESULTS

We analyzed 477 pregnancies. The main outcome occurred in 102/477 (21.4%) women. The final model included previous vaginal delivery (odds ratio [OR] 0.088; 95% confidence interval [CI] 0.04-0.21), height (OR 0.904; 95% CI 0.87-0.94), body mass index before delivery (OR 1.084; 95% CI 1.02-1.15), ultrasonographic estimated fetal weight (OR 3.965; 95% CI 2.18-7.22), and ultrasonographic cervical length (OR 1.065; 95% CI 1.04-1.09) as predictors. Area under the receiver operating characteristics curve was 0.826 (95% CI 0.78-0.87). For a 5% false-positive rate, the sensitivity, specificity, and positive and negative likelihood ratios were 44.1%, 94.9%, 8.7, and 0.59, respectively.

CONCLUSION

Our model combining clinical and ultrasonographic features might offer individualized counseling regarding risk of cesarean delivery to women who are candidates for IOL.