{"paper_id":"1dc65c9f-94b7-4dfa-b97d-9412dbbfaf23","body_text":"Enhancing VBAC Prediction with AI-Powered Temporal Dynamics: Integrating Decision Support into a Shared Decision-Making Platform for Intrapartum Care | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Enhancing VBAC Prediction with AI-Powered Temporal Dynamics: Integrating Decision Support into a Shared Decision-Making Platform for Intrapartum Care Ching-Fu Wang, Mu-En Lee, Cherng-Chia Yang, Shu-Wen Chen, Hsiang-Wei Hu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6188292/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Taiwan has a high caesarean section (CS) rate, ranging from 37% to 38%. Vaginal Birth After Cesarean (VBAC) offers a potential solution to reduce these rates. However, the prevalence of VBAC remains below 0.5%, primarily due to concerns about risks of adverse maternal and perinatal outcomes. Objectives: This study aims to evaluate the predictive performance of various machine learning (ML) models using pregnancy, labor, and intervention-related features to predict VBAC success and support real-time clinical decision-making during labor. Study Design: This retrospective exploratory study analyzed data collected from a hospital in northern Taiwan between January 2019 and May 2023. Statistical methods included demographic comparisons, feature evaluations, and model performance metrics such as accuracy, precision, recall, F1-score, and area under the curve (AUC). SHapley Additive exPlanations (SHAP) analysis was used to interpret feature importance and labor progression. Results: A comparison between the VBAC Failure group (n=22) and VBAC Success group (n=33), totaling 55 records from 36 pregnant women, revealed significant differences in parity, spontaneous rupture of membranes, cervical dilation (at both 0 cm and 10 cm), and labor progression slope. Models incorporating high-impact features demonstrated superior performance compared to those utilizing only pregnancy-related data. The Random Forest model achieved an accuracy of 94% and an AUC of 0.96 in predicting labor progression. SHAP analysis further identified key predictors across different stages of labor, including pregnancy-related features (body mass index, prior vaginal birth, maternal age), static features (spontaneous rupture of membranes, time since rupture), and dynamic features (cervical dilation and labor slope). Conclusion: This integrative approach, which combines clinical expertise with predictive analytics, provides clinicians with a valuable tool for real-time labor evaluation and decision-making. By offering more accurate predictions of labor progression, particularly in the context of VBAC, this approach has the potential to significantly improve maternal and neonatal outcomes Labor outcomes machine learning real-time decision-making VBAC prediction Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Background Taiwan has sustained high national caesarean rates, with more than 1 in 3 women experiencing caesarean section (CS) each year ( 1 ), over double the rate recommended by the World Health Organization ( 2 ). A study recommended that increasing VBAC (Vaginal Birth After Cesarean) in line with other countries could be an appropriate way to lower CS rates in Taiwan ( 3 ). However, the prevalence of VBAC has remained lower than 0.5% over two decades ( 1 ). Medical malpractice was most frequently cited by Taiwanese obstetricians reluctant to offer VBAC to women ( 4 – 6 ) because of concern about uterine rupture and increased risk of adverse maternal and neonatal outcomes ( 7 – 9 ). Artificial intelligence (AI) has emerged as a powerful tool in healthcare, offering the potential to support medical professionals and patients in making informed decisions ( 10 ). Based on previous research findings, we recently developed an AI predictive model for VBAC, which was integrated into an innovative decision-aid birth choice platform for shared decision-making (SDM) ( 11 , 12 ). Through this model, pregnant women can input seven crucial factors—maternal age, gravidity, body mass index (BMI), gestational week, history of previous vaginal birth, prolonged labor, gestational diabetes, and pregnancy-induced hypertension—to promptly ascertain their likelihood of achieving a successful VBAC during pregnancy ( 12 ). This AI calculator empowers SDM between healthcare providers and pregnant women by providing personalized, data-driven insights and risk assessments for VBAC ( 12 , 13 ). Intrapartum stages are full of instant changes, with several factors affecting the success rate of VBAC during labor ( 14 – 16 ). Labor characteristics, intrapartum interventions, and perinatal outcomes ultimately determine the success rates of VBAC ( 15 , 16 ). Despite numerous studies examining factors to achieve successful VBAC ( 17 , 18 ), few studies have explored the success rates of VBAC with AI prediction. This study aims to evaluate the predictive performance of various machine learning (ML) models by analyzing pregnancy, labor, and intervention-related features to predict VBAC success and facilitate real-time clinical decision-making during labor. 2. Methods 2.1 Study design This was a retrospective exploratory study. Figure 1 illustrated the architecture of an AI-powered VBAC prediction system designed for real-time use during the first stage of labor. Initially, the system linked to an AI-powered VBAC prediction system and inputted relevant VBAC features during the early phase of labor. Based on previous relevant studies, the features adopted include four types of parameters such as pregnancy features, static features in the stage of labor, dynamic features in the stage of labor, and intervention feature in the stage of labor fed into the AI model ( 19 , 20 ). The AI system then predicted the VBAC success rate during the intrapartum stage. This prediction assisted clinicians in making informed decisions regarding the likelihood of a successful VBAC during labor. The system was accessible on various devices, including personal computers, tablets, and mobile phones, ensuring its usability in diverse clinical environments. This integrated approach aimed to enhance real-time decision-making, improving outcomes for both the mother and the baby. 2.2 Data sources and data collection This study was approved by the Institutional Review Board of Fu Jen Catholic University (No. C111042), ensuring adherence to all ethical guidelines and regulations throughout the research process. Data was collected from a regional hospital in northern Taiwan. This study presented a flowchart detailing the selection and outcomes of pregnant women analyzed for VBAC success in Fig. 2 . The initial cohort comprised 44 pregnant women, with 61 records collected between January 2019 and May 2023. After excluding 8 women due to missing data, the final cohort comprised 36 pregnant women with 55 records, representing 90.02% of the initial cohort. The outcomes for these 36 women were categorized into VBAC Success and VBAC Failure. The VBAC Success group included 22 women (40% of the final cohort), while the VBAC Failure group comprised 33 records (60%). The flowchart illustrated the data selection process and the distribution of VBAC outcomes within the study population. 2.3 ML models establishment In this study, we compared four ML models: Logistic Regression (LR) ( 21 ), Support Vector Machine (SVM) ( 22 ), Extreme Gradient Boosting (XGB) ( 23 ), and Random Forest (RF) ( 24 ) to evaluate their strengths and limitations. LR offered efficiency and interpretability but struggled with non-linearity ( 25 ). SVM supported non-linear modeling but requires longer training ( 26 , 27 ). XGB excelled in feature engineering and large datasets, while RF managed high-dimensional data and reduced overfitting but lacked interpretability ( 28 , 29 , 30 , 31 ). Performance was assessed using accuracy, precision, recall, and F1-score, with cross-validation ensuring reliability 2.4 Model performance evaluation We employed binary classification models using SHAP (Shapley Additive exPlanations) values for interpretable feature selection ( 27 ) ( 32 ). Four ML models—LR, SVM, XGB, and RF—were tested on a standardized dataset, representing different learning approaches. We used accuracy, precision, recall, F1-score, AUC and receiver operating characteristic (ROC) curve to assess performance, providing insights into the models' classification accuracy and handling of class imbalances. Additionally, SHAP values identified key predictive features and their influence on model decisions, while temporal SHAP analysis tracked cervical dilation changes ( 33 ). A 5-fold cross-validation ensured robustness and mitigated overfitting, enhancing model reliability ( 34 ). Combining performance metrics, SHAP-based feature selection, and temporal analysis allowed us to assess model effectiveness, identify critical predictive features, and enhance interpretability for a comprehensive understanding of binary classification tasks. 2.5 Statistical analysis This study analyzed demographic comparisons between the VBAC Success and Failure group, feature-based evaluations, model performance metrics, and interpretable insights through SHAP analysis, for VBAC prediction and the factors influencing labor outcomes. Descriptive statistics summarized the demographic characteristics, including maternal age, BMI, parity, previous vaginal birth, previous prolonged labor and other relevant clinical features. Statistical analysis was conducted on all features, categorized into pregnancy features, static features in the stage of labor, dynamic features in the stage of labor, and intervention features in the stage of labor. Differences between groups were analyzed using appropriate statistical tests, including t-tests for continuous variables and Fisher’s exact tests for categorical variables, to identify features significantly associated with VBAC outcomes. ML models were evaluated using accuracy, precision, recall, F1-score, and AUC, with cross-validation ensuring reliability. ( 12 ). SHAP values highlighted influential features, while temporal analysis provided insights into cervical dilation progression, improving the understanding of labor dynamics. 3. Results 3.1 Characteristic statistics In this study, we performed a descriptive statistical analysis comparing VBAC success (n = 33) and VBAC failure (n = 22) among 36 participants (Table 1 ). Regarding the pregnancy features, only parity showed a significant difference between the two groups ( p = 0.034), with the VBAC success group exhibiting a higher mean and broader distribution. In the static features in the stage of labor, only spontaneous rupture of membranes was significantly more Table 1 Characteristic statistics. VBAC Success (n = 33) VBAC Failure (n = 22) p -value Overall (n = 55) Pregnancy features Maternal age [y, [min-max]] 34.7 ± 3.5 [29–42] 33.7 ± 3.1 [28–39] 0.266 34.3 ± 3.3 [28–42] BMI [kg/m2, [min-max]] 26.9 ± 3.6 [22.1–35.3] 28.2 ± 3.9 [22.5–38.4] 0.225 27.4 ± 3.8 [22.1–38.4] Parity [n, [min-max]] 2.7 ± 1.0 [2–5] 2.2 ± 0.53 [2–4] 0.034* 2.5 ± 0.8 [2–5] Previous vaginal birth [n, (%)] 5 (15.1) 1 (4.5) 0.384 6 (10.9) Previous prolonged labor [n, (%)] 4 (12.1) 4 (18.1) 0.700 8 (14.5) Pregnancy Hypertension [n, (%)] 9 (29.0) 4 (18.1) 0.561 13 (24.5) Pregnancy Diabetes [n, (%)] 6 (19.4) 6 (27.3) 0.730 12 (22.6) Pregnancy Proteinuria [n, (%)] 7 (22.6) 3 (13.6) 0.647 10 (18.9) Static features in the stage of labor Admission rupture of membranes [n, (%)] 28 (84.8) 21 (95.4) 0.384 49 (89.0) Spontaneous rupture of membranes [n, (%)] 17 (51.5) 1 (4.5) 0.0003*** 18 (32.7) Intrapartum Proteinuria [n, (%)] 5 (15.2) 1 (4.5) 0.384 6 (10.9) Intrapartum Diabetes [n, (%)] 0 (0) 1 (4.5) 0.400 1 (1.8) Intrapartum Hypertension [n, (%)] 2 (6.1) 4 (18.2) 0.204 6 (10.9) Time elapsed since the membrane of rupture [min, [min-max]] 123.0 ± 147.6 [4-570] 64.3 ± 154.5 [1-456] 0.166 99.5 ± 151.7 [1-570] Heartrate [BPM, [min-max]] 136.5 ± 10.7 [105–150] 140.5 ± 9.0 [131–175] 0.185 138.2 ± 10.1 [105–175] Dynamic features in the stage of labor The first stage of labor Dilation = 0.0 cm [n, %] 0 (0.0) 16 (72.7) < 0.0001*** 16 (29.1) Dilation = 2.0 cm [n, %] 4 (12.1) 1 (4.5) 0.637 5 (9.1) Dilation = 4.0 cm [n, %] 5 (15.1) 2 (9.1) 0.689 7 (12.7) Dilation = 6.0 cm [n, %] 6 (18.2) 1 (4.5) 0.222 7 (12.7) Dilation = 8.0 cm [n, %] 5 (15.1) 0 (0.0) 0.075 5 (9.1) Dilation = 10.0 cm [n, %] 13 (39.4) 2 (9.1) 0.029* 15 (27.3) Slope [cm/min] 0.08 ± 0.1 [0.0-0.6] 0.003 ± 0.009 [0.0-0.04] 0.001** 0.05 ± 0.1 [0.0-0.6] Intervention feature in the stage of labor Induction [n, (%)] 2 (6.0) 1 (4.5) 1.000 3 (5.4) p -value: * p < 0.05, ** p < 0.01, *** p < 0.001 frequent in the VBAC success group (51.5% vs. 4.5%, p = 0.0003). For dynamic features in the stage of labor, cervical dilation at 0.0 cm and 10.0 cm showed significant differences ( p < 0.0001 and p = 0.029), with a higher percentage of VBAC failure at 0.0 cm (72.7% vs. 0.0%) and a lower percentage at 10.0 cm (9.1% vs. 39.4%) compared to the VBAC success group. Additionally, the slope feature was significantly different ( p = 0.001). For intervention feature in the stage of labor, induction showed no significant difference ( p = 1.000). These results underscore important associations between maternal factors, labor characteristics, and interventions with VBAC outcomes, providing valuable insights for clinical decision-making. 3.2 The model performance comparison between every ML model This study evaluated the performance of RF, SVM, XGB, and LR models using different feature sets to predict pregnancy and labor outcomes. In Table 2 , The analyzed feature sets included: ( 1 ) pregnancy features only, ( 2 ) pregnancy and static features during labor, and ( 3 ) pregnancy and all features in the stage of labor for model comparison. Four ML model performed all the best with pregnancy and all features in the stage with accuracy, precision, recall, F1-score and AUC. Moreover, Fig. 3 showed ROC curves comparing different feature sets across all model types. High-impact features (SHAP ≥ 0.02) were identified from the best-performing model, which used the RF architecture with pregnancy and all features in the stage of labor (Table 2 ). High-impact features identified were BMI (SHAP = 0.05), previous vaginal birth (SHAP = 0.02), maternal age (SHAP = 0.02), spontaneous rupture of membranes (SHAP = 0.02), time elapsed since the membrane of rupture (SHAP = 0.13), dilation (SHAP = 0.06), and slope (SHAP = 0.11). Table 3 shows that the RF model achieved the highest accuracy (0.94 ± 0.09), precision (1.00 ± 0.00), recall (0.74 ± 0.22) and F1-score (0.92 ± 0.10) among the evaluated models, indicating its superior performance in accurately identifying positive cases. However, SVM model demonstrated the highest AUC values (0.97 ± 0.05), slightly exceeding the AUC of RF model (AUC = 0.96 ± 0.06), suggesting similar effectiveness in distinguishing between positive and negative classes. Figure 4 provided a ROC comparison of the models using high-impact features, illustrating the RF model as the most effective based on the evaluated metrics. Table 2 Performance comparison of the model using different features across all model types, including RF, SVM XGB, and LR models. Model type with different features Accuracy Precision Recall F1-score AUC Random Forest (RF) Pregnancy feature model 0.50 ± 0.15 0.59 ± 0.07 0.75 ± 0.22 0.65 ± 0.13 0.40 ± 0.20 Pregnancy and static features in the stage of labor model 0.80 ± 0.07 0.85 ± 0.13 0.90 ± 0.12 0.86 ± 0.03 0.84 ± 0.18 Pregnancy and all features in the stage of labor model 0.85 ± 0.15 0.85 ± 0.18 0.87 ± 0.17 0.85 ± 0.15 0.92 ± 0.16 Support Vector Machine (SVM) Pregnancy feature model 0.63 ± 0.07 0.65 ± 0.03 0.95 ± 0.10 0.77 ± 0.05 0.40 ± 0.09 Pregnancy and static features in the stage of labor model 0.60 ± 0.08 0.64 ± 0.03 0.90 ± 0.12 0.75 ± 0.07 0.55 ± 0.20 Pregnancy and all features in the stage of labor model 0.79 ± 0.04 0.80 ± 0.16 0.83 ± 0.15 0.79 ± 0.05 0.92 ± 0.07 eXtreme Gradient Boosting (XGB) Pregnancy feature model 0.63 ± 0.12 0.73 ± 0.05 0.70 ± 0.19 0.71 ± 0.12 0.57 ± 0.16 Pregnancy and static features in the stage of labor model 0.77 ± 0.13 0.83 ± 0.14 0.85 ± 0.12 0.83 ± 0.09 0.84 ± 0.15 Pregnancy and all features in the stage of labor model 0.85 ± 0.05 0.89 ± 0.14 0.83 ± 0.15 0.84 ± 0.05 0.84 ± 0.14 Logistic Regression (LR) Pregnancy feature model 0.67 ± 0.11 0.68 ± 0.07 0.95 ± 0.10 0.79 ± 0.07 0.59 ± 0.24 Pregnancy and static features in the stage of labor model 0.70 ± 0.07 0.71 ± 0.06 0.95 ± 0.10 0.81 ± 0.04 0.69 ± 0.27 Pregnancy and all features in the stage of labor model 0.81 ± 0.06 0.84 ± 0.13 0.79 ± 0.18 0.79 ± 0.08 0.89 ± 0.10 Table 3 Performance comparison of different model types using high-impact features. Model type Accuracy Precision Recall F1-score AUC RF 0.94 ± 0.09 1.00 ± 0.00 0.87 ± 0.17 0.92 ± 0.10 0.96 ± 0.06 SVM 0.87 ± 0.08 0.92 ± 0.10 0.84 ± 0.20 0.86 ± 0.10 0.97 ± 0.05 XGB 0.80 ± 0.11 0.91 ± 0.17 0.74 ± 0.22 0.78 ± 0.11 0.85 ± 0.08 LR 0.83 ± 0.09 0.88 ± 0.15 0.79 ± 0.18 0.81 ± 0.10 0.93 ± 0.07 3.3 Model Explainability in Predicting Labor Progression Figure 5 illustrated cervix dilation progress and feature importance, showing how predictive factors change dynamically throughout labor. The study used trained models and SHAP values to highlight the shifting importance of features in predicting labor outcomes, providing insights for clinical decision-making. Key high-impact features included pregnancy features, static features in the stage of labor, and dynamic features in the stage of labor, aligning with significant results in Table 1 . At 0 cm dilation, dynamic features (dilation and slope) were crucial for determining the timing of delivery, especially if a cesarean section was considered. From 2 cm to 4 cm dilation, the importance of dynamic features decreased, while static features become more critical. Between 2 cm and 8 cm dilation except for the 6 cm, the time elapsed since the membrane of rupture became the most significant indicator, marking rapid cervical dilation. At 6 and 10 cm dilation, the focus shifts to overall parameters similar to the 0 cm dilation, to assess labor progression. The features were ranked according to their impact on the model's output, with dilation consistently being the most significant predictor across different stages. The visualization effectively combines clinical insights with data-driven predictions, highlighting the dynamic changes and key factors influencing labor progression and outcomes. This integrated approach aids in understanding the critical elements involved in predicting labor progression, providing valuable information for clinicians and expectant mothers. 4. Discussion This study evaluated the predictive performance of different ML models using various feature sets to forecast labor outcomes. It was observed that models incorporating all features did not necessarily yield the best predictive power. Instead, models using high-impact features ( 35 , 36 )—those combining pregnancy period parameters, static feature in the stage of labor, and dynamic features in the stage of labor—demonstrated superior performance compared to models relying solely on pregnancy-related features. The RF model was the top performer in predicting labor progression, achieving 94% accuracy. SHAP value analysis showed how model interpretations evolved across dilation stages. Cervical dilation in 0 cm, the early latent phase, indicated higher vaginal birth failure risk with slow progression, highlighting the importance of dynamic dilation features. In the late latent phase (2 to 4 cm), dynamic features decreased, and feature distribution became more balanced, reflecting a transitional stage before entering active labor. In the active phase (6 to 10 cm), successful vaginal birth became more likely, with continuous dilation and slope posing high risk. The model's focused on dynamic features during these phases aligns with clinical observations, providing valuable insights for timely interventions ( 37 , 38 ). Figure 5 effectively integrates clinical insights with predictive analytics, illustrating key factors influencing labor outcomes and enhancing understanding for clinicians and expectant mothers. This integrated approach enhances the understanding of labor progression, offering valuable insights for clinicians and expectant mothers. This study’s sample size was sufficient, as ML quality often outweighs quantity. Despite a small cohort, SHAP analysis improved interpretability, ensuring robustness. The RF model’s 94% accuracy suggests adequate data for training ( 39 , 40 ). While larger datasets enhanced generalizability, this cohort (86.89% of the initial group) minimizes data loss and maintains clinical significance. Balanced representation between VBAC Success and VBAC Failure groups ensured meaningful learning. Given obstetrics' data collection challenges, this approach remains valid with advanced methods like SHAP ( 40 , 41 ), offering a foundation for future research. Contrasted with previously published models ( 12 ) focused on pregnancy features, which were suitable for use by pregnant individuals, this predictive platform integrated labor-relevant parameters, enhancing real-time evaluation. This supported physicians in natural childbirth care, improving decision-making and outcomes. The integration of ML models into VBAC prediction offered a valuable decision-support tool for obstetricians during labor by incorporating dynamic labor features such as cervical dilation, labor progression slope, etc. This approach could reduce unnecessary repeat cesarean sections while ensuring maternal and neonatal safety. Additionally, the use of SHAP analysis provided interpretable insights into predictive factors, bridging the gap between clinical expertise and data-driven decision-making. Implementing such models in clinical workflows could standardize labor management, offering a more personalized and evidence-based approach to childbirth. Further validation using larger, multi-center datasets was needed to enhance generalizability. Future research should refine model architecture, integrate tools into electronic health records (EHRs) and address ethical considerations such as patient consent and clinician trust. Evaluating ML-driven decision support’s impact on clinical outcomes and workflow efficiency will provide deeper insights into its benefits and limitations. Strengths and limitations Despite promising results, this study has limitations. First, the dataset, though comprehensive, may not fully represent all populations, affecting generalizability. Differences in healthcare practices and demographics could impact model performance. Second, while SHAP values improve interpretability, they cannot eliminate potential biases from feature selection or data limitations. Validating findings across diverse datasets is essential. Third, this study focused on high predictive accuracy without addressing clinical integration. Practical used requires user-friendly interfaces and seamless EHRs integration. Lastly, ethical and legal concerns, such as patient consent and data privacy, must be considered for safe implementation. Ongoing research was needed to enhance model reliability and clinical applicability in obstetric care. 5. Conclusion This study demonstrated the efficacy of ML models in predicting labor outcomes by integrating diverse clinical parameters beyond traditional pregnancy-related features. Our findings indicate that focusing on high-impact features—specifically pregnancy period parameters, static and dynamic labor parameters—significantly enhances predictive performance. The RF model proved most accurate, with SHAP values enhancing interpretability across labor stages. Visual insights bridge clinical expertise with predictive analytics, aiding real-time decision-making. By incorporating dynamic labor features, our model provides a robust tool for improving childbirth assessments and interventions. This research advances obstetrics by leveraging ML to enhance labor monitoring, ultimately optimizing maternal and neonatal health outcomes. Abbreviations VBAC: Vaginal Birth After Cesarean CS: Caesarean Section AI: Artificial Intelligence SDM: Shared Decision-Making ML: Machine Learning SHAP: SHapley Additive exPlanations AUC: Area Under the Curve ROC: Receiver Operating Characteristic BMI: Body Mass Index LR: Logistic Regression SVM: Support Vector Machine XGB: Extreme Gradient Boosting RF: Random Forest EHRs: Electronic Health Records Declarations Availability of data and material The datasets during the current study are available from the corresponding author on reasonable request. Acknowledgements We would like to express our sincere gratitude to Saint Paul’s Hospital for providing the valuable data necessary for this research. Funding This work was supported by the Ministry of Science and Technology (grant number MOST 110-2314-B-227-006-MY2). Author information Shu-Wen Chen and Hsiang-Wei Hu contributed equally to this work. Authors and Affiliations National Taipei University of Nursing and Health Science, School of Nursing, Taipei, Taiwan Shu-Wen Chen & Pei-Hung Liao School of Biomedical Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei City 11031, Taiwan Ching-Fu Wang Department of Information and Communications Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan Hsiang-Wei Hu Saint Paul’s Hospital, Department of Obstetrics and Gynecology, Taoyuan, Taiwan Cherng-Chia Yang National Yang Ming Chiao Tung University, College of Medicine, School of Medicine, Taipei, Taiwan Mu-En Lee National Taipei University of Nursing and Health Science, Smart Healthcare Interdisciplinary College, Taipei, Taiwan Chao-Yang Kuo Contributions SWC and HWH conceived the study design; CFW, SWC, and HWH drafted the manuscript; MEL and CCY were responsible for the methodology; CFW, MEL, and HWH conducted the validation; CYK and PHL were involved in the formal analysis; SWC was responsible for the investigation; CCY, SWC, and HWH were involved in data curation. The manuscript was reviewed and approved by CFW and HWH, and supervision was provided by SWC. All the authors have approved the submitted version, and have agreed both to be personally accountable for the author's own contributions and to ensure that questions related to the accuracy or integrity of any part of the work. Corresponding author Correspondence to Shu-Wen Chen and Hsiang-Wei Hu. Ethics declarations Ethics approval and consent to participate This study was conducted under the approval of the Ethics Committee of Fu Jen Catholic University (REC number: C111042), adhering strictly to the Helsinki Declaration guidelines and regulations prior to commencement. As this was retrospective medical record data, the requirement for obtaining written informed consent from all participants was waived by the Ethics Committee of Fu Jen Catholic University. Consent for publication Not applicable. Competing interests The authors declare no competing interests. References Health Promotion Administration. 2022 Statistics of birth reporting system. Taiwan: Ministry of Health and Welfare; 2023. Betrán AP, Torloni MR, Zhang JJ, Gülmezoglu AM, Aleem HA, Carroli G, WHO Working Group on Caesareans Section. WHO statement on caesarean section rates. BJOG. Int J Obstetr Gynaecol. 2016;123(5):667–70. Tsai HT, Wu CH. Vaginal birth after cesarean section-The world trend and local experience in Taiwan. Taiwan J Obstet Gynecol. 2017;56(1):41–5. Chen SW, Bucknall T, Hutchinson A, Nagle C. Decision-making on birth choices following a caesarean delivery in Taiwan. Australia: Deakin University; 2015. Chen SW. Birth Choice Decision Aids in Women Planning Vaginal Birth after Cesarean Delivery. Formos J Med. 2021;25(4):460–73. Chen SW, Cheng PJ. Mode of birth following a primary caesarean section: Taiwanese obstetricians' decision-making strategies. Taiwan Midwives J. 2015(57):55–69. İzbudak G, Tozkır E, Cogendez E, Uzun F, Eser SK. Comparison of maternal-neonatal results of vaginal birth after cesarean and elective repeat cesarean delivery. Ginekologia polska. 2021;92(4):306–11. Akter MJ, Shirin E. The Outcome of Vaginal Birth After One Caesarean Section (VBAC). J Bangladesh Coll Physicians Surg. 2021;39(1). Deshmukh U, Denoble AE, Son M. Trial of labor after cesarean, vaginal birth after cesarean, and the risk of uterine rupture: an expert review. Am J Obstet Gynecol. 2024;230(3):S783–803. Younis HA, Eisa TAE, Nasser M, Sahib TM, Noor AA, Alyasiri OM, et al. A systematic review and meta-analysis of artificial intelligence tools in medicine and healthcare: applications, considerations, limitations, motivation and challenges. Diagnostics. 2024;14(1):109. Chen SW, Shorten A, Yeh CC, Kao CH, Lu YY, Hu HW. An innovative web-based decision-aid about birth after cesarean for shared decision making in Taiwan: study protocol for a randomized control trial. Trials. 2023;24(1):103. Yang CC, Wang CF, Lin WM, Chen SW, Hu HW. Evaluating the performance of an AI-powered VBAC prediction system within a decision-aid birth choice platform for shared decision-making. Digit Health. 2024;10:20552076241257014. Chu J, Keedle H, Sutcliffe K, Blumenthal N, Levett K. The outcomes for women planning a VBAC at a private hospital in Australia. Birth. 2024. Li H, Yang L, Peng J, Cheng W, Ma H, Wu S, et al. Duration time of labor progression for pregnant women of vaginal birth after cesarean in Hubei, China. Ir J Med Sci (1971-). 2024;193(3):1351–8. Lan Y, Pan S, Chen B, Peng L, Chen R, Hua Y, et al. Labor characteristics and intrapartum interventions in women with vaginal birth after cesarean section. BMC Pregnancy Childbirth. 2022;22(1):589. Doğan EK, Çevik S. Determining the Impact of Cervical Dilation at Admission on Intrapartum Interventions and Labor Satisfaction in Pregnant Women. J Educ Res Nursing/Hemşirelikte Eğitim ve Araştırma Dergisi. 2024;21(2). Wu Y, Kataria Y, Wang Z, Ming W-K, Ellervik C. Factors associated with successful vaginal birth after a cesarean section: a systematic review and meta-analysis. BMC Pregnancy Childbirth. 2019;19:1–12. Hassan H, Zedan H, Said D. Successful Vaginal Birth Subsequent to Ceasarean Section. Mansoura Nurs J. 2024;11(1):343–55. Rao L, Gao W, Fu K, Zhang Y, Huang Y, Zhou X, et al. A prediction model for successful vaginal birth after caesarean delivery in Chinese mothers by using machine learning. Int J Syst Control Inform Process. 2024;4(2):123–37. Khan W, Zaki N, Ghenimi N, Ahmad A, Bian J, Masud MM, et al. Predicting preterm birth using explainable machine learning in a prospective cohort of nulliparous and multiparous pregnant women. PLoS ONE. 2023;18(12):e0293925. Thagaard IN, Stentoft-Larsen V, Iglesias MT, Demharter S, Krebs L. Prediction of vaginal birth after previous cesarean delivery using explainable machine learning models. 2024. Valkenborg D, Rousseau A-J, Geubbelmans M, Burzykowski T. Support vector machines. Am J Orthod Dentofac Orthop. 2023;164(5):754–7. Ali ZA, Abduljabbar ZH, Taher HA, Sallow AB, Almufti SM. Exploring the power of eXtreme gradient boosting algorithm in machine learning: A review. Acad J Nawroz Univ. 2023;12(2):320–34. Hu J, Szymczak S. A review on longitudinal data analysis with random forest. Brief Bioinform. 2023;24(2):bbad002. Lu Y, Duong T, Miao Z, Thieu T, Lamichhane J, Ahmed A, et al. A novel hyperparameter search approach for accuracy and simplicity in disease prediction risk scoring. J Am Med Inform Assoc. 2024;31(8):1763–73. Azzeh M, Elsheikh Y, Nassif AB, Angelis L. Examining the performance of kernel methods for software defect prediction based on support vector machine. Sci Comput Program. 2023;226:102916. Modhugu VR, Ponnusamy S. Comparative Analysis of Machine Learning Algorithms for Liver Disease Prediction: SVM, Logistic Regression, and Decision Tree. Asian J Res Comput Sci. 2024;17(6):188–201. Tarwidi D, Pudjaprasetya SR, Adytia D, Apri M. An optimized XGBoost-based machine learning method for predicting wave run-up on a sloping beach. MethodsX. 2023;10:102119. Yang J-Y, Hu HW, Liu C-H, Chen K-Y, Un C-H, Huang C-C, et al. editors. Differencing time series as an important feature extraction for intradialytic hypotension prediction using machine learning. 2021 IEEE 3rd Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability (ECBIOS); 2021: IEEE. Somi S, Jubair S, Cooper D, Wang P. XGSleeve: detecting sleeve incidents in well completion by using XGBoost classifier. Front Artif Intell. 2023;6:1243584. Yun H, Choi J, Park JH. Prediction of critical care outcome for adult patients presenting to emergency department using initial triage information: an XGBoost algorithm analysis. JMIR Med Inf. 2021;9(9):e30770. Santos MR, Guedes A, Sanchez-Gendriz I. SHapley Additive exPlanations (SHAP) for Efficient Feature Selection in Rolling Bearing Fault Diagnosis. Mach Learn Knowl Extr. 2024;6(1):316–41. Li R, Feng K, An T, Cheng P, Wei L, Zhao Z, et al. Enhanced Insights into Effluent Prediction in Wastewater Treatment Plants: Comprehensive Deep Learning Model Explanation Based on SHAP. ACS ES&T Water. 2024;4(4):1904–15. Bates S, Hastie T, Tibshirani R. Cross-validation: what does it estimate and how well does it do it? J Am Stat Assoc. 2024;119(546):1434–45. Lipschuetz M, Guedalia J, Rottenstreich A, Persky MN, Cohen SM, Kabiri D, et al. Prediction of vaginal birth after cesarean deliveries using machine learning. Am J Obstet Gynecol. 2020;222(6):613. e1-. e12. Awawdeh S, Rawashdeh H, Aljalodi H, Alshorman S. Vaginal birth after cesarean section prediction model for Jordanian population. Comput Biol Chem. 2023;104:107877. Tesfahun TD, Awoke AM, Kefale MM, Balcha WF, Nega AT, Gezahegn TW, et al. Factors associated with successful vaginal birth after one lower uterine transverse cesarean section delivery. Sci Rep. 2023;13(1):8871. Addisu D, Gebeyehu NA, Biru S, Belachew YY. Vaginal birth after cesarean section and its associated factors in Ethiopia: a systematic review and meta-analysis. Sci Rep. 2023;13(1):7882. Ben Yehuda O, Itelman E, Vaisman A, Segal G, Lerner B. Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study. J Med Internet Res. 2024;26:e48595. Dong T, Sinha S, Zhai B, Fudulu D, Chan J, Narayan P, et al. Performance drift in machine learning models for cardiac surgery risk prediction: retrospective analysis. JMIRx Med. 2024;5(1):e45973. Huang E-H, Hu H-W, Jheng W-L, Chen K-Y, Liu C-H, Chi H-Y, et al. editors. Feature Selection for Intradialytic Blood Pressure Value Prediction Using GRU-based Method Under RFECV algorithm. 2021 9th International Conference on Orange Technology (ICOT); 2021: IEEE. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-6188292\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":428182464,\"identity\":\"7360f69f-c65e-4cf6-b807-ef23ba37e140\",\"order_by\":0,\"name\":\"Ching-Fu Wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Taipei Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Ching-Fu\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":428182465,\"identity\":\"3722c983-56a2-4d16-a9db-843b5c263805\",\"order_by\":1,\"name\":\"Mu-En Lee\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Yang Ming Chiao Tung University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mu-En\",\"middleName\":\"\",\"lastName\":\"Lee\",\"suffix\":\"\"},{\"id\":428182469,\"identity\":\"d4d8eb50-4e32-4dff-a797-f3d93a7e2421\",\"order_by\":2,\"name\":\"Cherng-Chia Yang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Saint Paul’s Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Cherng-Chia\",\"middleName\":\"\",\"lastName\":\"Yang\",\"suffix\":\"\"},{\"id\":428182471,\"identity\":\"8c6004d7-613c-4733-84eb-204f205c38dc\",\"order_by\":3,\"name\":\"Shu-Wen Chen\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAz0lEQVRIiWNgGAWjYDACCTBpAyLYQARjA5Fa0hh4EFqYidJymAQt8rN7zCQ+7jifZy+Re+zhDwYb2Q0H+I9J4NNicOeMmeTMM7eLeSTy0o15GNKMNxxgZsOvRSLH7DZv2+3EHiBDGujCRJCWG3gdNgOo5W/bObAWyR8M/wlrYbgB1MLYdgCsRYKH4QBhLQY30sp/9rYlJ/aceWMmzWOQbDzzMLP5D/wOS95s8LPNLrG9HeSwCjvZvuONjw3wOgzNUiAmFJOjYBSMglEwCggDADQcR9ZRFR61AAAAAElFTkSuQmCC\",\"orcid\":\"\",\"institution\":\"National Taipei University of Nursing and Health Science\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Shu-Wen\",\"middleName\":\"\",\"lastName\":\"Chen\",\"suffix\":\"\"},{\"id\":428182474,\"identity\":\"f9f19250-4ca5-4f47-b28a-0e4c649e6d00\",\"order_by\":4,\"name\":\"Hsiang-Wei Hu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Industrial Technology Research Institute\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hsiang-Wei\",\"middleName\":\"\",\"lastName\":\"Hu\",\"suffix\":\"\"},{\"id\":428182475,\"identity\":\"68c3a53e-42fd-4470-b771-b09bad65b7ff\",\"order_by\":5,\"name\":\"Chao-Yang Kuo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Taipei University of Nursing and Health Science\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Chao-Yang\",\"middleName\":\"\",\"lastName\":\"Kuo\",\"suffix\":\"\"},{\"id\":428182477,\"identity\":\"a8092cf5-0c30-42d2-a490-cc11952a6526\",\"order_by\":6,\"name\":\"Pei-Hung Liao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Taipei University of Nursing and Health Science\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Pei-Hung\",\"middleName\":\"\",\"lastName\":\"Liao\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-03-09 11:08:19\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-6188292/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-6188292/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":78733972,\"identity\":\"bb6854fb-1395-4733-8209-5b3162b52b31\",\"added_by\":\"auto\",\"created_at\":\"2025-03-18 07:55:48\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":324182,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe architecture of a real-time prediction system based on an AI-powered VBAC model for immediate VBAC during the first stage of labor (1st Labor).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/5f17b9a5ae558134a3c895e3.png\"},{\"id\":78733973,\"identity\":\"7b6745ce-ec4f-4eb7-9a71-fe0a762f4f01\",\"added_by\":\"auto\",\"created_at\":\"2025-03-18 07:55:48\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":67222,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eRecruitment Flowchart for Pregnant Women with VBAC Success and Failure during 1st Labor.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/06368a67ca4551192c7471d3.png\"},{\"id\":78734743,\"identity\":\"2c8069ec-0e30-4f53-a191-bd77ada660f2\",\"added_by\":\"auto\",\"created_at\":\"2025-03-18 08:03:48\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":176302,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eROC comparison of the model using different features across all model types, including LR, SVM XGB, and RF models.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/0f66e4cbef205057fde58e7d.png\"},{\"id\":78734745,\"identity\":\"bd41c933-0afa-43d8-b9ed-5468ad210efb\",\"added_by\":\"auto\",\"created_at\":\"2025-03-18 08:03:49\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":106732,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eROC comparison of the model type using high-impact features.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/6cb81fab4e201cec01d66369.png\"},{\"id\":78734738,\"identity\":\"b2f62e42-1b59-4c74-a555-1cdbcf613bd0\",\"added_by\":\"auto\",\"created_at\":\"2025-03-18 08:03:48\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":151024,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eSHAP values for high-impact features (Pregnancy features, Static features in the stage of labor, and Dynamic features during labor) in relation to cervical dilation progression in the first stage of labor (from 0 cm to 10 cm).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/cb5740092e72feba4cfa2f90.png\"},{\"id\":100237286,\"identity\":\"1fe64cf9-8d15-411b-9efd-12027ffc096b\",\"added_by\":\"auto\",\"created_at\":\"2026-01-14 12:41:27\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1610058,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6188292/v1/9d5590a7-86a1-446d-8756-6fff9733c67b.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Enhancing VBAC Prediction with AI-Powered Temporal Dynamics: Integrating Decision Support into a Shared Decision-Making Platform for Intrapartum Care\",\"fulltext\":[{\"header\":\"1. Background\",\"content\":\"\\u003cp\\u003eTaiwan has sustained high national caesarean rates, with more than 1 in 3 women experiencing caesarean section (CS) each year (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e), over double the rate recommended by the World Health Organization (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e). A study recommended that increasing VBAC (Vaginal Birth After Cesarean) in line with other countries could be an appropriate way to lower CS rates in Taiwan (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e). However, the prevalence of VBAC has remained lower than 0.5% over two decades (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). Medical malpractice was most frequently cited by Taiwanese obstetricians reluctant to offer VBAC to women (\\u003cspan additionalcitationids=\\\"CR5\\\" citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e) because of concern about uterine rupture and increased risk of adverse maternal and neonatal outcomes (\\u003cspan additionalcitationids=\\\"CR8\\\" citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eArtificial intelligence (AI) has emerged as a powerful tool in healthcare, offering the potential to support medical professionals and patients in making informed decisions (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e). Based on previous research findings, we recently developed an AI predictive model for VBAC, which was integrated into an innovative decision-aid birth choice platform for shared decision-making (SDM) (\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). Through this model, pregnant women can input seven crucial factors\\u0026mdash;maternal age, gravidity, body mass index (BMI), gestational week, history of previous vaginal birth, prolonged labor, gestational diabetes, and pregnancy-induced hypertension\\u0026mdash;to promptly ascertain their likelihood of achieving a successful VBAC during pregnancy (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). This AI calculator empowers SDM between healthcare providers and pregnant women by providing personalized, data-driven insights and risk assessments for VBAC (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eIntrapartum stages are full of instant changes, with several factors affecting the success rate of VBAC during labor (\\u003cspan additionalcitationids=\\\"CR15\\\" citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). Labor characteristics, intrapartum interventions, and perinatal outcomes ultimately determine the success rates of VBAC (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). Despite numerous studies examining factors to achieve successful VBAC (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e), few studies have explored the success rates of VBAC with AI prediction. This study aims to evaluate the predictive performance of various machine learning (ML) models by analyzing pregnancy, labor, and intervention-related features to predict VBAC success and facilitate real-time clinical decision-making during labor.\\u003c/p\\u003e\"},{\"header\":\"2. Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.1 Study design\\u003c/h2\\u003e \\u003cp\\u003eThis was a retrospective exploratory study. Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e illustrated the architecture of an AI-powered VBAC prediction system designed for real-time use during the first stage of labor. Initially, the system linked to an AI-powered VBAC prediction system and inputted relevant VBAC features during the early phase of labor. Based on previous relevant studies, the features adopted include four types of parameters such as pregnancy features, static features in the stage of labor, dynamic features in the stage of labor, and intervention feature in the stage of labor fed into the AI model (\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). The AI system then predicted the VBAC success rate during the intrapartum stage. This prediction assisted clinicians in making informed decisions regarding the likelihood of a successful VBAC during labor. The system was accessible on various devices, including personal computers, tablets, and mobile phones, ensuring its usability in diverse clinical environments. This integrated approach aimed to enhance real-time decision-making, improving outcomes for both the mother and the baby.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.2 Data sources and data collection\\u003c/h2\\u003e \\u003cp\\u003eThis study was approved by the Institutional Review Board of Fu Jen Catholic University (No. C111042), ensuring adherence to all ethical guidelines and regulations throughout the research process. Data was collected from a regional hospital in northern Taiwan. This study presented a flowchart detailing the selection and outcomes of pregnant women analyzed for VBAC success in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e. The initial cohort comprised 44 pregnant women, with 61 records collected between January 2019 and May 2023. After excluding 8 women due to missing data, the final cohort comprised 36 pregnant women with 55 records, representing 90.02% of the initial cohort. The outcomes for these 36 women were categorized into VBAC Success and VBAC Failure. The VBAC Success group included 22 women (40% of the final cohort), while the VBAC Failure group comprised 33 records (60%). The flowchart illustrated the data selection process and the distribution of VBAC outcomes within the study population.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.3 ML models establishment\\u003c/h2\\u003e \\u003cp\\u003eIn this study, we compared four ML models: Logistic Regression (LR) (\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e), Support Vector Machine (SVM) (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e), Extreme Gradient Boosting (XGB) (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e), and Random Forest (RF) (\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e) to evaluate their strengths and limitations. LR offered efficiency and interpretability but struggled with non-linearity (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e). SVM supported non-linear modeling but requires longer training (\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e). XGB excelled in feature engineering and large datasets, while RF managed high-dimensional data and reduced overfitting but lacked interpretability (\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e). Performance was assessed using accuracy, precision, recall, and F1-score, with cross-validation ensuring reliability\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec6\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.4 Model performance evaluation\\u003c/h2\\u003e \\u003cp\\u003eWe employed binary classification models using SHAP (Shapley Additive exPlanations) values for interpretable feature selection (\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e) (\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e). Four ML models\\u0026mdash;LR, SVM, XGB, and RF\\u0026mdash;were tested on a standardized dataset, representing different learning approaches. We used accuracy, precision, recall, F1-score, AUC and receiver operating characteristic (ROC) curve to assess performance, providing insights into the models' classification accuracy and handling of class imbalances. Additionally, SHAP values identified key predictive features and their influence on model decisions, while temporal SHAP analysis tracked cervical dilation changes (\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e). A 5-fold cross-validation ensured robustness and mitigated overfitting, enhancing model reliability (\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e). Combining performance metrics, SHAP-based feature selection, and temporal analysis allowed us to assess model effectiveness, identify critical predictive features, and enhance interpretability for a comprehensive understanding of binary classification tasks.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.5 Statistical analysis\\u003c/h2\\u003e \\u003cp\\u003eThis study analyzed demographic comparisons between the VBAC Success and Failure group, feature-based evaluations, model performance metrics, and interpretable insights through SHAP analysis, for VBAC prediction and the factors influencing labor outcomes. Descriptive statistics summarized the demographic characteristics, including maternal age, BMI, parity, previous vaginal birth, previous prolonged labor and other relevant clinical features. Statistical analysis was conducted on all features, categorized into pregnancy features, static features in the stage of labor, dynamic features in the stage of labor, and intervention features in the stage of labor. Differences between groups were analyzed using appropriate statistical tests, including t-tests for continuous variables and Fisher\\u0026rsquo;s exact tests for categorical variables, to identify features significantly associated with VBAC outcomes. ML models were evaluated using accuracy, precision, recall, F1-score, and AUC, with cross-validation ensuring reliability. (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). SHAP values highlighted influential features, while temporal analysis provided insights into cervical dilation progression, improving the understanding of labor dynamics.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.1 Characteristic statistics\\u003c/h2\\u003e \\u003cp\\u003eIn this study, we performed a descriptive statistical analysis comparing VBAC success (n\\u0026thinsp;=\\u0026thinsp;33) and VBAC failure (n\\u0026thinsp;=\\u0026thinsp;22) among 36 participants (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). Regarding the pregnancy features, only parity showed a significant difference between the two groups (\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;0.034), with the VBAC success group exhibiting a higher mean and broader distribution. In the static features in the stage of labor, only spontaneous rupture of membranes was significantly more\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003eCharacteristic statistics.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"6\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eVBAC Success (n\\u0026thinsp;=\\u0026thinsp;33)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003eVBAC Failure (n\\u0026thinsp;=\\u0026thinsp;22)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e\\u003cem\\u003ep\\u003c/em\\u003e-value\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eOverall (n\\u0026thinsp;=\\u0026thinsp;55)\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy features\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eMaternal age [y, [min-max]]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e34.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.5 [29\\u0026ndash;42]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e33.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.1 [28\\u0026ndash;39]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.266\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e34.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.3 [28\\u0026ndash;42]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBMI [kg/m2, [min-max]]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e26.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.6 [22.1\\u0026ndash;35.3]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e28.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.9 [22.5\\u0026ndash;38.4]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.225\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e27.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.8 [22.1\\u0026ndash;38.4]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eParity [n, [min-max]]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e2.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.0 [2\\u0026ndash;5]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e2.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.53 [2\\u0026ndash;4]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.034*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e2.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.8 [2\\u0026ndash;5]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePrevious vaginal birth [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5 (15.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.384\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (10.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePrevious prolonged labor [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e4 (12.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e4 (18.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.700\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e8 (14.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy Hypertension [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e9 (29.0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e4 (18.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.561\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e13 (24.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy Diabetes [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6 (19.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e6 (27.3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.730\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e12 (22.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy Proteinuria [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e7 (22.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e3 (13.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.647\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e10 (18.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eStatic features in the stage of labor\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAdmission rupture of membranes [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e28 (84.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e21 (95.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.384\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e49 (89.0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSpontaneous rupture of membranes [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e17 (51.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.0003***\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e18 (32.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eIntrapartum Proteinuria [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5 (15.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.384\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (10.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eIntrapartum Diabetes [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0 (0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.400\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e1 (1.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eIntrapartum Hypertension [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e2 (6.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e4 (18.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.204\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e6 (10.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTime elapsed since the membrane of rupture [min, [min-max]]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e123.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;147.6 [4-570]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e64.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;154.5 [1-456]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.166\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e99.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;151.7 [1-570]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eHeartrate [BPM, [min-max]]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e136.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7 [105\\u0026ndash;150]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e140.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.0 [131\\u0026ndash;175]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.185\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e138.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1 [105\\u0026ndash;175]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDynamic features in the stage of labor\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"5\\\" nameend=\\\"c6\\\" namest=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eThe first stage of labor\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;0.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0 (0.0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e16 (72.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.0001***\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e16 (29.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;2.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e4 (12.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.637\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e5 (9.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;4.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5 (15.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e2 (9.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.689\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e7 (12.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;6.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6 (18.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.222\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e7 (12.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;8.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5 (15.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e0 (0.0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.075\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e5 (9.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eDilation\\u0026thinsp;=\\u0026thinsp;10.0 cm [n, %]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e13 (39.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e2 (9.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.029*\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e15 (27.3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSlope [cm/min]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.08\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1 [0.0-0.6]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c4\\\" namest=\\\"c3\\\"\\u003e \\u003cp\\u003e0.003\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.009 [0.0-0.04]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.001**\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.05\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1 [0.0-0.6]\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eIntervention feature in the stage of labor\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eInduction [n, (%)]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c3\\\" namest=\\\"c2\\\"\\u003e \\u003cp\\u003e2 (6.0)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1 (4.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1.000\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e3 (5.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003e\\u003cem\\u003ep\\u003c/em\\u003e-value: *\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05, **\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01, ***\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003efrequent in the VBAC success group (51.5% vs. 4.5%, \\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;0.0003). For dynamic features in the stage of labor, cervical dilation at 0.0 cm and 10.0 cm showed significant differences (\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.0001 and \\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;0.029), with a higher percentage of VBAC failure at 0.0 cm (72.7% vs. 0.0%) and a lower percentage at 10.0 cm (9.1% vs. 39.4%) compared to the VBAC success group. Additionally, the slope feature was significantly different (\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;0.001). For intervention feature in the stage of labor, induction showed no significant difference (\\u003cem\\u003ep\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.000). These results underscore important associations between maternal factors, labor characteristics, and interventions with VBAC outcomes, providing valuable insights for clinical decision-making.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec10\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.2 The model performance comparison between every ML model\\u003c/h2\\u003e \\u003cp\\u003eThis study evaluated the performance of RF, SVM, XGB, and LR models using different feature sets to predict pregnancy and labor outcomes. In Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e, The analyzed feature sets included: (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e) pregnancy features only, (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e) pregnancy and static features during labor, and (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e) pregnancy and all features in the stage of labor for model comparison. Four ML model performed all the best with pregnancy and all features in the stage with accuracy, precision, recall, F1-score and AUC. Moreover, Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e showed ROC curves comparing different feature sets across all model types.\\u003c/p\\u003e \\u003cp\\u003eHigh-impact features (SHAP\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.02) were identified from the best-performing model, which used the RF architecture with pregnancy and all features in the stage of labor (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). High-impact features identified were BMI (SHAP\\u0026thinsp;=\\u0026thinsp;0.05), previous vaginal birth (SHAP\\u0026thinsp;=\\u0026thinsp;0.02), maternal age (SHAP\\u0026thinsp;=\\u0026thinsp;0.02), spontaneous rupture of membranes (SHAP\\u0026thinsp;=\\u0026thinsp;0.02), time elapsed since the membrane of rupture (SHAP\\u0026thinsp;=\\u0026thinsp;0.13), dilation (SHAP\\u0026thinsp;=\\u0026thinsp;0.06), and slope (SHAP\\u0026thinsp;=\\u0026thinsp;0.11). Table\\u0026nbsp;\\u003cspan refid=\\\"Tab3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e shows that the RF model achieved the highest accuracy (0.94\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.09), precision (1.00\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.00), recall (0.74\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.22) and F1-score (0.92\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10) among the evaluated models, indicating its superior performance in accurately identifying positive cases. However, SVM model demonstrated the highest AUC values (0.97\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05), slightly exceeding the AUC of RF model (AUC\\u0026thinsp;=\\u0026thinsp;0.96\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06), suggesting similar effectiveness in distinguishing between positive and negative classes. Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e provided a ROC comparison of the models using high-impact features, illustrating the RF model as the most effective based on the evaluated metrics.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab2\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 2\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003ePerformance comparison of the model using different features across all model types, including RF, SVM XGB, and LR models.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"6\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eModel type with different features\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eAccuracy\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003ePrecision\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eRecall\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eF1-score\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eAUC\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eRandom Forest (RF)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy feature model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.50\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.59\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.75\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.22\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.65\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.13\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.40\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.20\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and static features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.80\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.13\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.90\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.12\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.86\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.03\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.18\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and all features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.18\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.87\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.92\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eSupport Vector Machine (SVM)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy feature model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.63\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.65\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.03\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.95\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.77\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.40\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and static features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.60\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.08\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.64\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.03\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.90\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.12\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.75\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.55\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.20\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and all features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.04\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.80\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.92\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eeXtreme Gradient Boosting (XGB)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy feature model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.63\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.12\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.73\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.70\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.19\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.71\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.12\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.57\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and static features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.77\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.13\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.14\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.12\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and all features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.89\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.14\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.14\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colspan=\\\"6\\\" nameend=\\\"c6\\\" namest=\\\"c1\\\"\\u003e \\u003cp\\u003eLogistic Regression (LR)\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy feature model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.67\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.11\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.68\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.95\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.59\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.24\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and static features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.70\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.71\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.95\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.81\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.04\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.69\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.27\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePregnancy and all features in the stage of labor model\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.81\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.13\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.18\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.08\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.89\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab3\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 3\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003ePerformance comparison of different model types using high-impact features.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"6\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eModel type\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eAccuracy\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003ePrecision\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eRecall\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eF1-score\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eAUC\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eRF\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.94\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.00\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.00\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.87\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.92\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.96\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSVM\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.87\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.08\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.92\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.20\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.86\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.97\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.05\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eXGB\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.80\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.11\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.91\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.74\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.22\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.78\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.11\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.85\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.08\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLR\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.83\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.88\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.15\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.79\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.18\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.81\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003e0.93\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.07\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.3 Model Explainability in Predicting Labor Progression\\u003c/h2\\u003e \\u003cp\\u003eFigure \\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e illustrated cervix dilation progress and feature importance, showing how predictive factors change dynamically throughout labor. The study used trained models and SHAP values to highlight the shifting importance of features in predicting labor outcomes, providing insights for clinical decision-making. Key high-impact features included pregnancy features, static features in the stage of labor, and dynamic features in the stage of labor, aligning with significant results in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e. At 0 cm dilation, dynamic features (dilation and slope) were crucial for determining the timing of delivery, especially if a cesarean section was considered. From 2 cm to 4 cm dilation, the importance of dynamic features decreased, while static features become more critical. Between 2 cm and 8 cm dilation except for the 6 cm, the time elapsed since the membrane of rupture became the most significant indicator, marking rapid cervical dilation. At 6 and 10 cm dilation, the focus shifts to overall parameters similar to the 0 cm dilation, to assess labor progression. The features were ranked according to their impact on the model's output, with dilation consistently being the most significant predictor across different stages. The visualization effectively combines clinical insights with data-driven predictions, highlighting the dynamic changes and key factors influencing labor progression and outcomes. This integrated approach aids in understanding the critical elements involved in predicting labor progression, providing valuable information for clinicians and expectant mothers.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"4. Discussion\",\"content\":\"\\u003cp\\u003eThis study evaluated the predictive performance of different ML models using various feature sets to forecast labor outcomes. It was observed that models incorporating all features did not necessarily yield the best predictive power. Instead, models using high-impact features (\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e)\\u0026mdash;those combining pregnancy period parameters, static feature in the stage of labor, and dynamic features in the stage of labor\\u0026mdash;demonstrated superior performance compared to models relying solely on pregnancy-related features.\\u003c/p\\u003e \\u003cp\\u003eThe RF model was the top performer in predicting labor progression, achieving 94% accuracy. SHAP value analysis showed how model interpretations evolved across dilation stages. Cervical dilation in 0 cm, the early latent phase, indicated higher vaginal birth failure risk with slow progression, highlighting the importance of dynamic dilation features. In the late latent phase (2 to 4 cm), dynamic features decreased, and feature distribution became more balanced, reflecting a transitional stage before entering active labor. In the active phase (6 to 10 cm), successful vaginal birth became more likely, with continuous dilation and slope posing high risk. The model's focused on dynamic features during these phases aligns with clinical observations, providing valuable insights for timely interventions (\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e). Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e effectively integrates clinical insights with predictive analytics, illustrating key factors influencing labor outcomes and enhancing understanding for clinicians and expectant mothers. This integrated approach enhances the understanding of labor progression, offering valuable insights for clinicians and expectant mothers.\\u003c/p\\u003e \\u003cp\\u003eThis study\\u0026rsquo;s sample size was sufficient, as ML quality often outweighs quantity. Despite a small cohort, SHAP analysis improved interpretability, ensuring robustness. The RF model\\u0026rsquo;s 94% accuracy suggests adequate data for training (\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). While larger datasets enhanced generalizability, this cohort (86.89% of the initial group) minimizes data loss and maintains clinical significance. Balanced representation between VBAC Success and VBAC Failure groups ensured meaningful learning. Given obstetrics' data collection challenges, this approach remains valid with advanced methods like SHAP (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e), offering a foundation for future research.\\u003c/p\\u003e \\u003cp\\u003eContrasted with previously published models (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e) focused on pregnancy features, which were suitable for use by pregnant individuals, this predictive platform integrated labor-relevant parameters, enhancing real-time evaluation. This supported physicians in natural childbirth care, improving decision-making and outcomes.\\u003c/p\\u003e \\u003cp\\u003eThe integration of ML models into VBAC prediction offered a valuable decision-support tool for obstetricians during labor by incorporating dynamic labor features such as cervical dilation, labor progression slope, etc. This approach could reduce unnecessary repeat cesarean sections while ensuring maternal and neonatal safety. Additionally, the use of SHAP analysis provided interpretable insights into predictive factors, bridging the gap between clinical expertise and data-driven decision-making. Implementing such models in clinical workflows could standardize labor management, offering a more personalized and evidence-based approach to childbirth.\\u003c/p\\u003e \\u003cp\\u003eFurther validation using larger, multi-center datasets was needed to enhance generalizability. Future research should refine model architecture, integrate tools into electronic health records (EHRs) and address ethical considerations such as patient consent and clinician trust. Evaluating ML-driven decision support\\u0026rsquo;s impact on clinical outcomes and workflow efficiency will provide deeper insights into its benefits and limitations.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eStrengths and limitations\\u003c/b\\u003e \\u003c/p\\u003e \\u003cp\\u003eDespite promising results, this study has limitations. First, the dataset, though comprehensive, may not fully represent all populations, affecting generalizability. Differences in healthcare practices and demographics could impact model performance. Second, while SHAP values improve interpretability, they cannot eliminate potential biases from feature selection or data limitations. Validating findings across diverse datasets is essential. Third, this study focused on high predictive accuracy without addressing clinical integration. Practical used requires user-friendly interfaces and seamless EHRs integration. Lastly, ethical and legal concerns, such as patient consent and data privacy, must be considered for safe implementation. Ongoing research was needed to enhance model reliability and clinical applicability in obstetric care.\\u003c/p\\u003e\"},{\"header\":\"5. Conclusion\",\"content\":\"\\u003cp\\u003eThis study demonstrated the efficacy of ML models in predicting labor outcomes by integrating diverse clinical parameters beyond traditional pregnancy-related features. Our findings indicate that focusing on high-impact features\\u0026mdash;specifically pregnancy period parameters, static and dynamic labor parameters\\u0026mdash;significantly enhances predictive performance. The RF model proved most accurate, with SHAP values enhancing interpretability across labor stages. Visual insights bridge clinical expertise with predictive analytics, aiding real-time decision-making. By incorporating dynamic labor features, our model provides a robust tool for improving childbirth assessments and interventions. This research advances obstetrics by leveraging ML to enhance labor monitoring, ultimately optimizing maternal and neonatal health outcomes.\\u003c/p\\u003e\"},{\"header\":\"Abbreviations\",\"content\":\"\\u003cp\\u003eVBAC: Vaginal Birth After Cesarean\\u003c/p\\u003e\\n\\u003cp\\u003eCS: Caesarean Section\\u003c/p\\u003e\\n\\u003cp\\u003eAI: Artificial Intelligence\\u003c/p\\u003e\\n\\u003cp\\u003eSDM: Shared Decision-Making\\u003c/p\\u003e\\n\\u003cp\\u003eML: Machine Learning\\u003c/p\\u003e\\n\\u003cp\\u003eSHAP: SHapley Additive exPlanations\\u003c/p\\u003e\\n\\u003cp\\u003eAUC: Area Under the Curve\\u003c/p\\u003e\\n\\u003cp\\u003eROC: Receiver Operating Characteristic\\u003c/p\\u003e\\n\\u003cp\\u003eBMI: Body Mass Index\\u003c/p\\u003e\\n\\u003cp\\u003eLR: Logistic Regression\\u003c/p\\u003e\\n\\u003cp\\u003eSVM: Support Vector Machine\\u003c/p\\u003e\\n\\u003cp\\u003eXGB: Extreme Gradient Boosting\\u003c/p\\u003e\\n\\u003cp\\u003eRF: Random Forest\\u003c/p\\u003e\\n\\u003cp\\u003eEHRs: Electronic Health Records\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and material\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe datasets during the current study are available from the corresponding author on reasonable request.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eWe would like to express our sincere gratitude to Saint Paul\\u0026rsquo;s Hospital for providing the valuable data necessary for this research.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis work was supported by the Ministry of Science and Technology (grant number MOST 110-2314-B-227-006-MY2).\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthor information\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eShu-Wen Chen and Hsiang-Wei Hu contributed equally to this work.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors and Affiliations\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNational Taipei University of Nursing and Health Science, School of Nursing, Taipei, Taiwan\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eShu-Wen Chen \\u0026amp; Pei-Hung\\u0026nbsp;Liao\\u003c/p\\u003e\\n\\u003cp\\u003eSchool of Biomedical Engineering, College of Biomedical Engineering, Taipei Medical University, Taipei City 11031, Taiwan\\u003c/p\\u003e\\n\\u003cp\\u003eChing-Fu Wang\\u003c/p\\u003e\\n\\u003cp\\u003eDepartment of Information and Communications Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan\\u003c/p\\u003e\\n\\u003cp\\u003eHsiang-Wei Hu\\u003c/p\\u003e\\n\\u003cp\\u003eSaint Paul\\u0026rsquo;s Hospital, Department of Obstetrics and Gynecology, Taoyuan, Taiwan\\u003c/p\\u003e\\n\\u003cp\\u003eCherng-Chia Yang\\u003c/p\\u003e\\n\\u003cp\\u003eNational Yang Ming Chiao Tung University, College of Medicine, School of Medicine, Taipei, Taiwan\\u003c/p\\u003e\\n\\u003cp\\u003eMu-En Lee\\u003c/p\\u003e\\n\\u003cp\\u003eNational Taipei University of Nursing and Health Science, Smart Healthcare Interdisciplinary College, Taipei, Taiwan\\u003c/p\\u003e\\n\\u003cp\\u003eChao-Yang Kuo\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eContributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eSWC and HWH conceived the study design; CFW, SWC, and HWH drafted the manuscript; MEL and CCY were responsible for the methodology; CFW, MEL, and HWH conducted the validation; CYK and PHL were involved in the formal analysis; SWC was responsible for the investigation; CCY, SWC, and HWH were involved in data curation. The manuscript was reviewed and approved by CFW and HWH, and supervision was provided by SWC. All the authors have approved the submitted version, and have agreed both to be personally accountable for the author\\u0026apos;s own contributions and to ensure that questions related to the accuracy or integrity of any part of the work.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCorresponding author\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eCorrespondence to Shu-Wen Chen and Hsiang-Wei Hu.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthics declarations\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthics approval and consent to participate\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was conducted under the approval of the Ethics Committee of Fu Jen Catholic University (REC number: C111042), adhering strictly to the Helsinki Declaration guidelines and regulations prior to commencement. As this was retrospective medical record data, the requirement for obtaining written informed consent from all participants was waived by the Ethics Committee of Fu Jen Catholic University.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare no competing interests.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eHealth Promotion Administration. 2022 Statistics of birth reporting system. Taiwan: Ministry of Health and Welfare; 2023.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBetr\\u0026aacute;n AP, Torloni MR, Zhang JJ, G\\u0026uuml;lmezoglu AM, Aleem HA, Carroli G, WHO Working Group on Caesareans Section. WHO statement on caesarean section rates. BJOG. Int J Obstetr Gynaecol. 2016;123(5):667\\u0026ndash;70.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTsai HT, Wu CH. Vaginal birth after cesarean section-The world trend and local experience in Taiwan. Taiwan J Obstet Gynecol. 2017;56(1):41\\u0026ndash;5.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen SW, Bucknall T, Hutchinson A, Nagle C. Decision-making on birth choices following a caesarean delivery in Taiwan. Australia: Deakin University; 2015.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen SW. Birth Choice Decision Aids in Women Planning Vaginal Birth after Cesarean Delivery. Formos J Med. 2021;25(4):460\\u0026ndash;73.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen SW, Cheng PJ. Mode of birth following a primary caesarean section: Taiwanese obstetricians' decision-making strategies. Taiwan Midwives J. 2015(57):55\\u0026ndash;69.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eİzbudak G, Tozkır E, Cogendez E, Uzun F, Eser SK. Comparison of maternal-neonatal results of vaginal birth after cesarean and elective repeat cesarean delivery. Ginekologia polska. 2021;92(4):306\\u0026ndash;11.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAkter MJ, Shirin E. The Outcome of Vaginal Birth After One Caesarean Section (VBAC). J Bangladesh Coll Physicians Surg. 2021;39(1).\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDeshmukh U, Denoble AE, Son M. Trial of labor after cesarean, vaginal birth after cesarean, and the risk of uterine rupture: an expert review. Am J Obstet Gynecol. 2024;230(3):S783\\u0026ndash;803.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYounis HA, Eisa TAE, Nasser M, Sahib TM, Noor AA, Alyasiri OM, et al. A systematic review and meta-analysis of artificial intelligence tools in medicine and healthcare: applications, considerations, limitations, motivation and challenges. Diagnostics. 2024;14(1):109.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen SW, Shorten A, Yeh CC, Kao CH, Lu YY, Hu HW. An innovative web-based decision-aid about birth after cesarean for shared decision making in Taiwan: study protocol for a randomized control trial. Trials. 2023;24(1):103.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYang CC, Wang CF, Lin WM, Chen SW, Hu HW. Evaluating the performance of an AI-powered VBAC prediction system within a decision-aid birth choice platform for shared decision-making. Digit Health. 2024;10:20552076241257014.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChu J, Keedle H, Sutcliffe K, Blumenthal N, Levett K. The outcomes for women planning a VBAC at a private hospital in Australia. Birth. 2024.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi H, Yang L, Peng J, Cheng W, Ma H, Wu S, et al. Duration time of labor progression for pregnant women of vaginal birth after cesarean in Hubei, China. Ir J Med Sci (1971-). 2024;193(3):1351\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLan Y, Pan S, Chen B, Peng L, Chen R, Hua Y, et al. Labor characteristics and intrapartum interventions in women with vaginal birth after cesarean section. BMC Pregnancy Childbirth. 2022;22(1):589.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDoğan EK, \\u0026Ccedil;evik S. Determining the Impact of Cervical Dilation at Admission on Intrapartum Interventions and Labor Satisfaction in Pregnant Women. J Educ Res Nursing/Hemşirelikte Eğitim ve Araştırma Dergisi. 2024;21(2).\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWu Y, Kataria Y, Wang Z, Ming W-K, Ellervik C. Factors associated with successful vaginal birth after a cesarean section: a systematic review and meta-analysis. BMC Pregnancy Childbirth. 2019;19:1\\u0026ndash;12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHassan H, Zedan H, Said D. Successful Vaginal Birth Subsequent to Ceasarean Section. Mansoura Nurs J. 2024;11(1):343\\u0026ndash;55.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRao L, Gao W, Fu K, Zhang Y, Huang Y, Zhou X, et al. A prediction model for successful vaginal birth after caesarean delivery in Chinese mothers by using machine learning. Int J Syst Control Inform Process. 2024;4(2):123\\u0026ndash;37.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKhan W, Zaki N, Ghenimi N, Ahmad A, Bian J, Masud MM, et al. Predicting preterm birth using explainable machine learning in a prospective cohort of nulliparous and multiparous pregnant women. PLoS ONE. 2023;18(12):e0293925.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eThagaard IN, Stentoft-Larsen V, Iglesias MT, Demharter S, Krebs L. Prediction of vaginal birth after previous cesarean delivery using explainable machine learning models. 2024.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eValkenborg D, Rousseau A-J, Geubbelmans M, Burzykowski T. Support vector machines. Am J Orthod Dentofac Orthop. 2023;164(5):754\\u0026ndash;7.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAli ZA, Abduljabbar ZH, Taher HA, Sallow AB, Almufti SM. Exploring the power of eXtreme gradient boosting algorithm in machine learning: A review. Acad J Nawroz Univ. 2023;12(2):320\\u0026ndash;34.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHu J, Szymczak S. A review on longitudinal data analysis with random forest. Brief Bioinform. 2023;24(2):bbad002.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLu Y, Duong T, Miao Z, Thieu T, Lamichhane J, Ahmed A, et al. A novel hyperparameter search approach for accuracy and simplicity in disease prediction risk scoring. J Am Med Inform Assoc. 2024;31(8):1763\\u0026ndash;73.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAzzeh M, Elsheikh Y, Nassif AB, Angelis L. Examining the performance of kernel methods for software defect prediction based on support vector machine. Sci Comput Program. 2023;226:102916.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eModhugu VR, Ponnusamy S. Comparative Analysis of Machine Learning Algorithms for Liver Disease Prediction: SVM, Logistic Regression, and Decision Tree. Asian J Res Comput Sci. 2024;17(6):188\\u0026ndash;201.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTarwidi D, Pudjaprasetya SR, Adytia D, Apri M. An optimized XGBoost-based machine learning method for predicting wave run-up on a sloping beach. MethodsX. 2023;10:102119.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYang J-Y, Hu HW, Liu C-H, Chen K-Y, Un C-H, Huang C-C, et al. editors. Differencing time series as an important feature extraction for intradialytic hypotension prediction using machine learning. 2021 IEEE 3rd Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability (ECBIOS); 2021: IEEE.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSomi S, Jubair S, Cooper D, Wang P. XGSleeve: detecting sleeve incidents in well completion by using XGBoost classifier. Front Artif Intell. 2023;6:1243584.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYun H, Choi J, Park JH. Prediction of critical care outcome for adult patients presenting to emergency department using initial triage information: an XGBoost algorithm analysis. JMIR Med Inf. 2021;9(9):e30770.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSantos MR, Guedes A, Sanchez-Gendriz I. SHapley Additive exPlanations (SHAP) for Efficient Feature Selection in Rolling Bearing Fault Diagnosis. Mach Learn Knowl Extr. 2024;6(1):316\\u0026ndash;41.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi R, Feng K, An T, Cheng P, Wei L, Zhao Z, et al. Enhanced Insights into Effluent Prediction in Wastewater Treatment Plants: Comprehensive Deep Learning Model Explanation Based on SHAP. ACS ES\\u0026amp;T Water. 2024;4(4):1904\\u0026ndash;15.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBates S, Hastie T, Tibshirani R. Cross-validation: what does it estimate and how well does it do it? J Am Stat Assoc. 2024;119(546):1434\\u0026ndash;45.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLipschuetz M, Guedalia J, Rottenstreich A, Persky MN, Cohen SM, Kabiri D, et al. Prediction of vaginal birth after cesarean deliveries using machine learning. Am J Obstet Gynecol. 2020;222(6):613. e1-. e12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAwawdeh S, Rawashdeh H, Aljalodi H, Alshorman S. Vaginal birth after cesarean section prediction model for Jordanian population. Comput Biol Chem. 2023;104:107877.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTesfahun TD, Awoke AM, Kefale MM, Balcha WF, Nega AT, Gezahegn TW, et al. Factors associated with successful vaginal birth after one lower uterine transverse cesarean section delivery. Sci Rep. 2023;13(1):8871.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAddisu D, Gebeyehu NA, Biru S, Belachew YY. Vaginal birth after cesarean section and its associated factors in Ethiopia: a systematic review and meta-analysis. Sci Rep. 2023;13(1):7882.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBen Yehuda O, Itelman E, Vaisman A, Segal G, Lerner B. Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study. J Med Internet Res. 2024;26:e48595.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDong T, Sinha S, Zhai B, Fudulu D, Chan J, Narayan P, et al. Performance drift in machine learning models for cardiac surgery risk prediction: retrospective analysis. JMIRx Med. 2024;5(1):e45973.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHuang E-H, Hu H-W, Jheng W-L, Chen K-Y, Liu C-H, Chi H-Y, et al. editors. Feature Selection for Intradialytic Blood Pressure Value Prediction Using GRU-based Method Under RFECV algorithm. 2021 9th International Conference on Orange Technology (ICOT); 2021: IEEE.\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Labor outcomes, machine learning, real-time decision-making, VBAC prediction\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-6188292/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-6188292/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eBackground:\\u003c/strong\\u003e Taiwan has a high caesarean section (CS) rate, ranging from 37% to 38%. Vaginal Birth After Cesarean (VBAC) offers a potential solution to reduce these rates. However, the prevalence of VBAC remains below 0.5%, primarily due to concerns about risks of adverse maternal and perinatal outcomes.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eObjectives:\\u003c/strong\\u003e This study aims to evaluate the predictive performance of various machine learning (ML) models using pregnancy, labor, and intervention-related features to predict VBAC success and support real-time clinical decision-making during labor.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eStudy Design: \\u003c/strong\\u003eThis retrospective exploratory study analyzed data collected from a hospital in northern Taiwan between January 2019 and May 2023. Statistical methods included demographic comparisons, feature evaluations, and model performance metrics such as accuracy, precision, recall, F1-score, and area under the curve (AUC). SHapley Additive exPlanations (SHAP) analysis was used to interpret feature importance and labor progression.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults: \\u003c/strong\\u003eA comparison between the VBAC Failure group (n=22) and VBAC Success group (n=33), totaling 55 records from 36 pregnant women, revealed significant differences in parity, spontaneous rupture of membranes, cervical dilation (at both 0 cm and 10 cm), and labor progression slope. Models incorporating high-impact features demonstrated superior performance compared to those utilizing only pregnancy-related data. The Random Forest model achieved an accuracy of 94% and an AUC of 0.96 in predicting labor progression. SHAP analysis further identified key predictors across different stages of labor, including pregnancy-related features (body mass index, prior vaginal birth, maternal age), static features (spontaneous rupture of membranes, time since rupture), and dynamic features (cervical dilation and labor slope).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusion:\\u003c/strong\\u003e This integrative approach, which combines clinical expertise with predictive analytics, provides clinicians with a valuable tool for real-time labor evaluation and decision-making. By offering more accurate predictions of labor progression, particularly in the context of VBAC, this approach has the potential to significantly improve maternal and neonatal outcomes\\u003c/p\\u003e\",\"manuscriptTitle\":\"Enhancing VBAC Prediction with AI-Powered Temporal Dynamics: Integrating Decision Support into a Shared Decision-Making Platform for Intrapartum Care\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-03-18 07:55:43\",\"doi\":\"10.21203/rs.3.rs-6188292/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"d0f0dfcb-39ad-4b84-b861-912b2b73b279\",\"owner\":[],\"postedDate\":\"March 18th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2026-01-14T12:40:16+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-03-18 07:55:43\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-6188292\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-6188292\",\"identity\":\"rs-6188292\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}