Mechanical Effects of a Maylard Scar During a Vaginal Birth After a Previous Caesarean

Patient-specific surrogate model to predict pelvic floor dynamics during vaginal delivery

Childbirth is a challenging event that can lead to long-term consequences such as prolapse or incontinence. While computational models are widely used to mimic vaginal delivery, their integration into clinical practice is hindered by time constraints. The primary goal of this study is to introduce an artificial intelligence pipeline that leverages patient-specific surrogate modeling to predict pelvic floor injuries during vaginal delivery. A finite element-based machine learning approach was implemented to generate a dataset with information from finite element simulations. Thousands of childbirth simulations were conducted, varying the dimensions of the pelvic floor muscles and the mechanical properties used for their characterization. Additionally, a mesh morphing algorithm was developed to obtain patient-specific models. Machine learning models, specifically tree-based algorithms such as Random Forest (RF) and Extreme Gradient Boosting, as well as Artificial Neural Networks, were trained to predict the nodal coordinates of nodes within the pelvic floor, aiming to predict the muscle stretch during a critical interval. The results indicate that the RF model performs best, with a mean absolute error (MAE) of 0.086 mm and a mean absolute percentage error of 0.38%. Overall, more than 80% of the nodes have an error smaller than 0.1 mm. The MAE for the calculated stretch is equal to 0.0011. The implemented pipeline allows loading the trained model and making predictions in less than 11 s. This work demonstrates the feasibility of implementing a machine learning framework in clinical practice to predict potential maternal injuries and assist in medical-decision making.

On the uncertainty quantification of the active uterine contraction during the second stage of labor simulation

Uterine contractions in the myometrium occur at multiple scales, spanning both organ and cellular levels. This complex biological process plays an essential role in the fetus delivery during the second stage of labor. Several finite element models of active uterine contractions have already been developed to simulate the descent of the fetus through the birth canal. However, the developed models suffer severe reliability issues due to the uncertain parameters. In this context, the present study aimed to perform the uncertainty quantification (UQ) of the active uterine contraction simulation to advance our understanding of pregnancy mechanisms with more reliable indicators. A uterus model with and without fetus was developed integrating a transversely isotropic Mooney-Rivlin material with two distinct fiber orientation architectures. Different contraction patterns with complex boundary conditions were designed and applied. A global sensitivity study was performed to select the most valuable parameters for the uncertainty quantification (UQ) process using a copula-based Monte Carlo method. As results, four critical material parameters (C 1 , C 2 , K , Ca 0 ) of the active uterine contraction model were identified and used for the UQ process. The stress distribution on the uterus during the fetus descent, considering first and second fiber orientation families, ranged from 0.144 to 1.234 MPa and 0.044 to 1.619 MPa, respectively. The simulation outcomes revealed also the segment-specific contraction pattern of the uterus tissue. The present study quantified, for the first time, the effect of uncertain parameters of the complex constitutive model of the active uterine contraction on the fetus descent process. As perspectives, a full maternal pelvis model will be coupled with reinforcement learning to automatically identify the delivery mechanism behind the cardinal movements of the fetus during the active expulsion process.

Uterine wound healing after caesarean section: A systematic review

The rate of caesarean section (CS) is increasing worldwide. Defects in uterine healing have a major gynaecological and obstetric impact (uterine rupture, caesarean scar defect, caesarean scar pregnancy, placenta accreta spectrum). The complex process of cellular uterine healing after surgery, and specifically after CS, remains poorly understood in contrast to skin wound healing. This literature review on uterine wound healing was mainly based on histological observations, particularly after CS. The primary objective of the review was to examine the effects of CS on uterine tissue at the cellular level, based on histological observations. The secondary objectives were to describe the biomechanical characteristics and the therapies used to improve scar tissue after CS. This review was performed using PRISMA criteria, and PubMed was the data source. The study included all clinical and animal model studies with CS and histological analysis of the uterine scar area (macroscopic, microscopic, immunohistochemical and biomechanical). Twenty studies were included: 10 human and 10 animal models. In total, 533 female humans and 511 female animals were included. Review articles, meeting abstracts, case series, case reports, and abstracts without access to full-text were excluded. The search was limited to studies published in English. No correlation was found between cutaneous and uterine healing. The histology of uterine scars is characterized by disorganized smooth muscle, fibrosis with collagen fibres and fewer endometrial glands. As for skin healing, the initial inflammation phase and mediation of some growth factors (particularly connective tissue growth factor, vascular endothelial growth factor, platelet-derived growth factor, tumour necrosis factor α and tumour necrosis factor β) seem to be essential. This initial phase has an impact on the subsequent phases of proliferation and maturation. Collagen appears to play a key role in the initial granulation tissue to replace the loss of substance. Subsequent maturation of the scar tissue is essential, with a decrease in collagen and smooth muscle restoration. Unlike skin, the glandular structure of uterine tissue could be responsible for the relatively high incidence of healing defects. Uterine scar defects after CS are characterized by an atrophic disorganized endometrium with atypia and a fibroblastic highly collagenic stromal reaction. Concerning immunohistochemistry, one study found a decrease in tumour necrosis factor β in uterine scar defects. No correlation was found between biomechanical characteristics (particularly uterine strength) and the presence of a collagenous scar after CS. Based on the findings of this review, an illustration of current understanding about uterine healing is provided. There is currently no validated prevention of caesarean scar defects. Various treatments to improve uterine healing after CS have been tested, and appeared to have good efficacy in animal studies: alpha lipoic acid, growth factors, collagen scaffolds and mesenchymal stem cells. Further prospective studies are needed.

Evaluating the performance of an AI-powered VBAC prediction system within a decision-aid birth choice platform for shared decision-making

Background

Vaginal birth after cesarean (VBAC) is generally regarded as a safe and viable birthing option for most women with prior cesarean delivery. Nonetheless, concerns about heightened risks of adverse maternal and perinatal outcomes have often dissuaded women from considering VBAC. This study aimed to assess the performance of an artificial intelligence (AI)-powered VBAC prediction system integrated into a decision-aid birth choice platform for shared decision-making (SDM).

Materials and Methods

Employing a retrospective design, we collected medical records from a regional hospital in northern Taiwan from January 2019 to May 2023. To explore a suitable model for tabular data, we compared two prevailing modeling approaches: tree-based models and logistic regression models. We subjected the tree-based algorithm, CatBoost, to binary classification.

Results

Forty pregnant women with 347 records were included. The CatBoost model demonstrated a robust performance, boasting an accuracy rate of 0.91 (95% confidence interval (CI): 0.86-0.94) and an area under the curve of 0.89 (95% CI: 0.86-0.93), surpassing both regression models and other boosting techniques. CatBoost captured the data characteristics on the significant impact of gravidity and the positive influence of previous vaginal birth, reinforcing established clinical guidelines, as substantiated by the SHapley Additive exPlanations analysis.

Conclusion

Using AI techniques offers a more accurate assessment of VBAC risks, boosting women's confidence in selecting VBAC as a viable birthing option. The seamless integration of AI prediction systems with SDM platforms holds a promising potential for enhancing the effectiveness of clinical applications in the domain of women's healthcare.