Emily Butler, Hanna Ellson, Hannah Taylor, Oliver Hugh, Jason Gardosi
{"title":"INTERGROWTH 21st versus GROW fundal height charts and antenatal detection of small-for-gestational age: A multicenter cohort study.","authors":"Emily Butler, Hanna Ellson, Hannah Taylor, Oliver Hugh, Jason Gardosi","doi":"10.1002/ijgo.71128","DOIUrl":"10.1002/ijgo.71128","url":null,"abstract":"","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"1058-1059"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674837/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148204739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Josien Westendorp, Rugiatu Jalloh, Sarrah Sheriff, Jenny Wussow, Guro Stamland, Sigrid-Anne Bjorøy, Francis Moses, Marcus J Rijken, Risa Hoffmann-Lonnee, Håkon A Bolkan, Alex J van Duinen
{"title":"A cross-sectional study of caesarean deliveries performed by medical doctors and associate clinicians in nine district and regional hospitals in Sierra Leone using the Robson Ten Group Classification System.","authors":"Josien Westendorp, Rugiatu Jalloh, Sarrah Sheriff, Jenny Wussow, Guro Stamland, Sigrid-Anne Bjorøy, Francis Moses, Marcus J Rijken, Risa Hoffmann-Lonnee, Håkon A Bolkan, Alex J van Duinen","doi":"10.1002/ijgo.71072","DOIUrl":"10.1002/ijgo.71072","url":null,"abstract":"<p><strong>Objective: </strong>Maternal and neonatal mortality remain high in Sierra Leone despite significant decline in the last decade. The aim of the present study was to evaluate obstetric care delivery using the Robson Ten Group Classification System (TGCS), focusing on caesarean delivery (CD) and task-sharing.</p><p><strong>Methods: </strong>This cross-sectional study analyzed maternity and surgical registry data from nine governmental hospitals, collected over 4 months in 2021 or 2022. Variables included mode and outcome of delivery, TGCS parameters, and provider cadre: associate clinician (AC) or medical doctor (MD).</p><p><strong>Results: </strong>Of 5748 women included in the study, 3806 (66.2%) delivered vaginally, 156 (2.7%) by assisted vaginal delivery and 1786 (31.1%) by CD. The largest TGCS groups were low risk primipara (28.8%) and multipara (47.0%). In the low risk primipara group, 450/1656 (27.2%) underwent CD; in the low risk multipara group, 604/2702 (22.4%) underwent CD. MDs performed 876 (49.1%) of the total CDs and ACs 830 (46.5%). ACs performed more CDs with the indication of twin pregnancy, previous CD, or malpresentation, whereas MDs performed more with the indication of breech, uterine rupture, and fetal distress.</p><p><strong>Conclusion: </strong>Facility-based CD rates in Sierra Leone are high, particularly among low risk primi- and multipara-Robson groups, while assisted vaginal delivery rates remain low. Nearly half of all CDs are performed by ACs, highlighting the importance of task-sharing in this context. Further in-depth analysis is warranted to understand the factors influencing CD decision making and the collaborative processes between ACs and MDs to improve maternal and neonatal outcomes.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"865-876"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674845/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148101565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Félicia Joinau-Zoulovits, Maëlle Silvestre De Sacy, Bruno Renevier, Sarah Abramowicz
{"title":"Risk factors for not proceeding to reconstructive surgery after female genital mutilation: A cohort study of 220 women in a specialized referral center.","authors":"Félicia Joinau-Zoulovits, Maëlle Silvestre De Sacy, Bruno Renevier, Sarah Abramowicz","doi":"10.1002/ijgo.71094","DOIUrl":"10.1002/ijgo.71094","url":null,"abstract":"<p><strong>Objective: </strong>This study identifies factors associated with not proceeding to clitoral reconstructive surgery among women with female genital mutilation (FGM) enrolled in a specialized surgical pathway.</p><p><strong>Methods: </strong>A retrospective cohort study was conducted at a multidisciplinary referral center in Montreuil, France, between January 2021 and December 2023. The study included 220 women who expressed a wish for reconstructive surgery during their initial assessment. Factors associated with non-progression to surgery were identified using univariable and multivariable logistic regression, supplemented by Bayesian analysis.</p><p><strong>Results: </strong>Of the 220 women, 35% did not proceed to surgery. In the multivariable model, lack of support within the care pathway (odds ratio [OR] 3.28, 95% confidence interval [CI] 1.46-7.55; P = 0.004) and financial vulnerability, specifically reliance on state welfare (OR 3.43, 95% CI 1.28-9.80; P = 0.017) or financial support from relatives/community (OR 6.18, 95% CI 1.40-28.9; P = 0.017), were independently associated with non-progression. Model discrimination was good (area under the curve 0.75). Bayesian analysis confirmed these findings (P[direction] >99%) and suggested that memory of the FGM event also played a role (OR 1.87, 95% credible interval 0.88-4.09; P[direction] 94.9%).</p><p><strong>Conclusion: </strong>Access to clitoral reconstruction is primarily determined by socioeconomic stability and social support rather than clinical factors alone. These findings highlight the need for multidisciplinary pathways that integrate social and psychological accompaniment to ensure equitable access to care for vulnerable women.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"846-856"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674851/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148042252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Augusto Rimoli Esteves, Gabriel de Castro Pereira Bittar, Bruna Marques Lopes, Sônia Maria Rolim Rosa Lima
{"title":"Sociodemographic determinants of sexuality among transmasculine and transfeminine individuals: A cross-sectional study in São Paulo, Brazil.","authors":"Augusto Rimoli Esteves, Gabriel de Castro Pereira Bittar, Bruna Marques Lopes, Sônia Maria Rolim Rosa Lima","doi":"10.1002/ijgo.71081","DOIUrl":"10.1002/ijgo.71081","url":null,"abstract":"<p><strong>Objective: </strong>This study assessed the associations between gender identity (transmasculine vs transfeminine) and sexuality-related variables, including marital status, sexual partners, sex work, and age at first intercourse.</p><p><strong>Methods: </strong>We conducted a cross-sectional retrospective study at a specialized transgender clinic in São Paulo, Brazil, analyzing medical records from March 2020 to August 2023. Data were routinely collected during clinical consultations through standardized questions for individualized care. Group assignment was based on self-identified gender identity documented in medical records. Chi-square and t-tests were applied (P < 0.05).</p><p><strong>Results: </strong>We evaluated 74 individuals (trans men = 49, trans women = 25). Trans women were more frequently single (76% vs 44.9%, P = 0.014). Most reported one to five partners, but trans women more frequently reported no partners (16.7%) or >15 partners (11.1%) (P = 0.043). Sex work was more common among trans women (21.7% vs 6.3%, P = 0.049). No significant difference in age at first intercourse was observed (P = 0.298).</p><p><strong>Conclusion: </strong>Gender identity significantly influences relationship status, sexual behavior, and sex work engagement among transgender individuals. Trans women experienced greater sexual vulnerability, reflecting social exclusion and structural transphobia. These findings underscore the need for inclusive sexual health strategies and policies reducing structural inequalities.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"756-761"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674826/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147959290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Identification of disproportionate reporting signals for ectopic pregnancy in the FAERS database from 2004 to 2024: A real-world pharmacovigilance study.","authors":"Lixia Fu, Jia Xu","doi":"10.1002/ijgo.71088","DOIUrl":"10.1002/ijgo.71088","url":null,"abstract":"<p><strong>Background: </strong>Ectopic pregnancy (EP) is a serious obstetric complication with a rising global incidence, partly related to the increasing use of hormonal therapies and assisted reproductive technologies. Although structural abnormalities and infectious factors have been extensively studied, the reporting patterns of EP involving specific drugs remain insufficiently characterized in large spontaneous reporting systems. Previous studies have been limited by small sample sizes and heterogeneous designs, and potential drug-related EP reporting signals have not been comprehensively evaluated.</p><p><strong>Methods: </strong>This retrospective real-world pharmacovigilance study used data from the FDA Adverse Event Reporting System (FAERS), covering reports submitted from the first quarter of 2004 (2004Q1) to the fourth quarter of 2024 (2024Q4). After deduplication, drug names were standardized using RxNorm and adverse event terms were standardized using MedDRA version 27.1. Reports involving EP-related coded terms were extracted. Disproportionality analysis using the reporting odds ratio and 95% confidence intervals was performed to evaluate drug-event reporting associations. Subgroup analyses were conducted according to drug class, indication, body weight, and time to onset. Weibull distribution modeling was applied to explore report-based temporal patterns of frequently reported drugs.</p><p><strong>Results: </strong>A total of 4924 EP reports were identified, predominantly among women of reproductive age, with 46.6% aged 24-35 years. Levonorgestrel, etonogestrel, and clomifene showed the strongest disproportionate reporting signals for EP. Sixteen drugs, including interferon beta-1a, mifepristone, and lamivudine, were identified as potential novel safety signals not described in product labeling. The distribution of signal-positive drugs varied across body weight strata, with montelukast and medroxyprogesterone showing notable reporting signals in higher-weight subgroups. Time-to-onset analysis identified both early failure and random failure reporting patterns, supporting the need for temporally tailored pharmacovigilance strategies.</p><p><strong>Conclusion: </strong>This study identified disproportionate reporting signals for EP in a large spontaneous reporting database. These findings should be interpreted as hypothesis-generating pharmacovigilance observations rather than evidence of incidence, causal effects, or clinically confirmed risk differences. The results might help prioritize drug-event pairs for further evaluation and warrant validation in well-designed epidemiologic and clinical studies.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"827-839"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148030483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xingchen Liu, Yu-Jun Xiong, Xiangda Meng, Tian Lv, Minyan Cai
{"title":"Mortality trends in US women with concurrent sepsis and urinary tract infection.","authors":"Xingchen Liu, Yu-Jun Xiong, Xiangda Meng, Tian Lv, Minyan Cai","doi":"10.1002/ijgo.71110","DOIUrl":"10.1002/ijgo.71110","url":null,"abstract":"<p><strong>Objective: </strong>Urinary tract infections (UTIs) are common among women and are a recognized source of sepsis, particularly among older adults. However, long-term national mortality trends among females with concurrent UTI and sepsis remain inadequately characterized.</p><p><strong>Methods: </strong>A retrospective population-based analysis was conducted using US mortality records from 1999 to 2023. Decedents were females aged ≥25 years with death certificates documenting both UTI (ICD-10: N39.0) and sepsis (A40-A41). Age-adjusted mortality rates (AAMRs) were computed using the 2000 US standard population. Joinpoint regression was applied to estimate annual percent change (APC) and average annual percent change (AAPC). Stratification included age, race, region, state, and urbanization level.</p><p><strong>Results: </strong>A total of 275 514 deaths were identified. The UTI-sepsis AAMR rose from 6.07 per 100 000 in 1999 to 9.65 per 100 000 in 2023 (AAPC 2.26%, 95% confidence interval [CI]: 1.68-2.84). A significant increase was observed from 1999 to 2014, with a slower subsequent rise and a notable acceleration after 2019. The highest mortality burden was observed among women aged ≥85 years, while the steepest relative increases occurred in middle-aged groups. The South exhibited the highest regional mortality. Non-Hispanic Black women showed persistently elevated rates. Non-metropolitan areas experienced higher and more rapidly increasing mortality than metropolitan areas.</p><p><strong>Conclusion: </strong>Mortality among US females with concurrent UTI and sepsis has increased over two decades, accompanied by widening age, racial, and geographic disparities. These trends underscore the need for targeted interventions.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"885-894"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148143270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dana Josephy, Yael Yagur, May Weinberg, Maya Ben-Zion, Merav Sharvit, Zvi Klein, Yair Daykan, Ron Schonman
{"title":"Comparative outcomes of abdominal wall endometriosis surgery: Specialized endometriosis surgeons versus general surgical teams.","authors":"Dana Josephy, Yael Yagur, May Weinberg, Maya Ben-Zion, Merav Sharvit, Zvi Klein, Yair Daykan, Ron Schonman","doi":"10.1002/ijgo.71085","DOIUrl":"10.1002/ijgo.71085","url":null,"abstract":"<p><strong>Objective: </strong>To assess the preoperative diagnostic accuracy of abdominal wall endometriosis (AWE) by general surgical teams and specialized endometriosis centers in assessing abdominal wall masses using clinical evaluation and imaging.</p><p><strong>Methods: </strong>A retrospective cohort study included patients with a pathologic diagnosis of AWE who underwent surgical excision between 2007 and 2022. Patients treated by specialized endometriosis surgeons were compared with those treated by general surgery teams. Data collected included patient demographics, lesion size, preoperative symptoms, diagnostic modalities, and surgical outcomes. Statistical analysis was conducted using univariant and multivariable regression models to identify factors associated with misdiagnosis. Correlation between the size of lesion and the need for the use of mesh was also done.</p><p><strong>Results: </strong>A total of 46 patients with AWE were included; 27 (58.7%) were treated by the specialized endometriosis center and 19 (41.3%) by general surgeons. Evaluation by a specialized endometriosis center demonstrated a significantly higher correct diagnosis rate compared to the general surgery group (96.3% vs 42.1%, P < 0.001), with no significant difference in complication rates. Regression model revealed that admission by general surgeons increased the likelihood of misdiagnosis by 28 times (28.01 95% confidence interval [CI]: 2.74-286.47, P = 0.005). Computed tomography (CT) and magnetic resonance imaging were used more frequently in the general surgery group 42.1% versus 3.7%, P = 0.001. Endometriosis ultrasound was used more frequently in the specialized endometriosis group (100% vs 0%, P < 0.001). Independent of the group the probability of mesh use increased markedly for lesions larger than 3 cm with near certainty for lesions larger than 5 cm.</p><p><strong>Conclusion: </strong>Evaluation by specialized endometriosis surgeons was associated with significantly higher diagnostic accuracy of AWE detection compared with evaluation by general surgical teams. Early referral to experienced endometriosis units may reduce misdiagnosis and improve surgical planning.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"857-864"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674816/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148054436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fiona Cody, Orla Franklin, Caoimhe Lynch, Zara Molphy, Patrick Dicker, Fergal Malone, Fionnuala M Breathnach
{"title":"Evaluation of a novel fetal echocardiography training programme in two tertiary care obstetric Centres.","authors":"Fiona Cody, Orla Franklin, Caoimhe Lynch, Zara Molphy, Patrick Dicker, Fergal Malone, Fionnuala M Breathnach","doi":"10.1002/ijgo.71091","DOIUrl":"10.1002/ijgo.71091","url":null,"abstract":"<p><strong>Objectives: </strong>Improving and maintaining high detection rates for major congenital heart disease (CHD) is a priority for successful prenatal anatomy screening programmes. The primary objective of this study was to evaluate the utility of on-site multidimensional targeted training in fetal cardiac screening.</p><p><strong>Methods: </strong>A prospective study evaluating a targeted fetal echocardiography (FE) training programme in two obstetric tertiary care units was established. The programme was designed and approved by a dedicated fetal cardiology team. The one-day intervention involved supervised hands-on sonography and a prescribed educational programme for sonographers and doctors. The programme was completed by 23 trainees. Evaluation of the programme involved pre and post-training questionnaires and retrospective evaluation of cardiac images from 10 randomly-selected anatomy screening examinations performed by each trainee before and after training. Images were assessed for both technical quality and 6 priority elements of cardiac screening taught in the programme. The results were described with summary statistics and non-parametric tests.</p><p><strong>Results: </strong>1850 fetal cardiac images obtained before and after training by 21/23 trainees were assessed. Significant improvement was seen in the priority cardiac views obtained, for example from 24%(5/21) to 86%(18/21) for bifurcating pulmonary artery (P-value 0.001), from 33%(7/21) to 86%(17/21) for differential atrio-ventricular valve insertion (P-value 0.001), In addition, technical image quality improvements were observed following training as well as sonographer confidence.</p><p><strong>Conclusion: </strong>A targeted FE training programme can improve understanding of the priority elements required to exclude major CHD in prenatal screening, improve image quality, and sonographer confidence in cardiac screening.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":"771-778"},"PeriodicalIF":2.7,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13674832/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147972371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Chromosomal aberrations of reproductive dysfunction: An in-depth retrospective study of 1943 individuals with unexplained infertility.","authors":"Farzaneh Iravani, Khosro Namjoo, Fateme Montazeri, Mohammadreza Dehghani, Leila Zanbagh, Seyed Mehdi Kalantar","doi":"10.1002/ijgo.71462","DOIUrl":"https://doi.org/10.1002/ijgo.71462","url":null,"abstract":"<p><strong>Objective: </strong>Fertility refers to the ability to achieve a clinically successful pregnancy. Chromosomal aberrations constitute a fundamental etiological factor in reproductive dysfunction, with prevalence rates substantially elevated in infertile populations compared to general demographic baselines. This comprehensive retrospective investigation aimed to elucidate the cytogenetic signatures underlying infertility pathogenesis through systematic karyotypic analysis.</p><p><strong>Methods: </strong>This retrospective study evaluated individuals referred to the Yazd Reproductive Sciences Institute between April 2016 and March 2025 for reproductive health concerns. Among 7768 individuals assessed during the study period, 1943 had infertility and underwent cytogenetic evaluation. Peripheral blood lymphocytes were analyzed using standardized Giemsa (GTG)-banding karyotyping at a resolution of 400-550 bands per haploid genome. Chromosomal abnormalities were classified according to the International System for Human Cytogenomic Nomenclature (ISCN) 2020 guidelines, encompassing structural variants, numerical abnormalities, complex karyotypic alterations, and polymorphic heteromorphisms. The demographic characteristics of the patients were systematically evaluated and analyzed in correlation with their chromosomal abnormalities.</p><p><strong>Results: </strong>Among the 1943 individuals with infertility, chromosomal abnormalities were identified in 229 cases (12.0%), including numerical sex chromosome abnormalities (46.7%), chromosomal polymorphic variants (41.5%), structural abnormalities (9.2%), and complex karyotypes (2.6%). Males accounted for 83.8% of abnormal cases, with Klinefelter syndrome representing the most frequent abnormality.</p><p><strong>Conclusion: </strong>This study highlights that detecting chromosomal abnormalities in infertile couples is crucial for elucidating infertility etiology align with optimizing treatment and genetic counseling.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149286678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Integrating large language models and expert knowledge to identify risk factors and predict mode of delivery with machine learning.","authors":"Farzaneh Rashidi, Saba Qanbarzadeh, Azam Orooji","doi":"10.1002/ijgo.71376","DOIUrl":"https://doi.org/10.1002/ijgo.71376","url":null,"abstract":"<p><strong>Background: </strong>Accurately predicting the mode of delivery is a significant challenge in obstetrics, with implications for maternal healthcare planning, resource allocation, and personalized patient counseling. While machine learning offers promising predictive capabilities, the selection of relevant clinical and socioeconomic risk factors from vast possibilities remains a critical and nontrivial step that directly affects model performance and clinical utility.</p><p><strong>Objectives: </strong>This study aimed to (1) predict the mode of delivery (vaginal vs. cesarean) using machine learning algorithms and (2) identify the key risk factors influencing delivery mode prediction.</p><p><strong>Methods: </strong>A retrospective study was conducted on eligible pregnant women who delivered at Bent-Al-Huda Hospital in Bojnourd, Iran. A hybrid feature identification approach was employed: large language models (LLMs) initially proposed 63 potential risk factors. Through iterative refinement involving domain experts and assessment of data availability and frequency, this list was consolidated to 35 final variables for modeling. Machine learning algorithms, including multilayer perceptron (MLP), decision tree, naïve Bayes, and support vector machine (SVM), were applied to build predictive models.</p><p><strong>Results: </strong>Among all models, SVM showed the best performance using all features, achieving an accuracy of 94.8% and an area under the curve (AUC) of 0.94. Decision tree analysis identified previous cesarean section history, insurance type, and fetal presentation as the three most important predictive factors. After excluding 40 women with prior cesarean delivery, SVM maintained robust performance with accuracy of 94.3% and an AUC of 0.92. Using antenatal features only, MLP achieved the highest accuracy before and after excluding women with previous cesarean section (92.9% and 92.2%), while SVM also showed the highest discriminative ability (accuracy 92.4%, AUC 0.90 and accuracy 91.3%, AUC 0.87, respectively).</p><p><strong>Conclusion: </strong>This study demonstrates that combining expert knowledge with LLMs enables effective risk factor identification and that careful feature selection allows the proposed model to predict delivery type with high accuracy.</p>","PeriodicalId":14164,"journal":{"name":"International Journal of Gynecology & Obstetrics","volume":" ","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149286765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}