Cardiotocography (CTG) is a widely used, cost-effective non-invasive fetal monitor, yet its predictive potential for adverse pregnancy outcomes is understudied. This study constructed an AI model based on CTG signals to estimate fetal biological age (CTGage), and validated CTGage-gap (CTGage minus true gestational age) as a novel digital risk biomarker. Based on 61,140 CTG signals from 11,385 pregnancies, a distribution-aligned augmented 1D residual CNN (Net1d) was trained for CTGage prediction. Subjects were split into five CTGage-gap subgroups, and severe underestimation and overestimation subgroups were pooled as a high-risk cohort for outcome incidence comparison. The average absolute error of the model was 10.86 days. The incidence of preterm birth and gestational diabetes mellitus (GDM) in the overestimation group increased, while in the underestimation group, there were increased risks of low birthweight, hypertensive disorders of pregnancy (HDP), and maternal anemia (all p < 0.05). The AI-driven CTGage can be used as a feasible non-invasive biomarker for predicting adverse pregnancy outcomes.
Researchers and clinicians are increasingly interested in the potential of AI to improve women’s health. Here, we argue that systemic harms can arise when AI is uncritically embraced as a revolution for women’s health. We focus on three intersecting concerns: (1) impoverished understandings of gender and sex, which are (2) fed uncritically into AI models surrounded by hype, leading to (3) harms rather than improvements in health. We call on health advocates and practitioners to critically examine what AI can and cannot contribute to women’s health.
Statins reduce cardiovascular risk but modestly increase type 2 diabetes (T2D). Evidence is limited by underrepresentation of women and lack of sex-specific analyses. In an umbrella review and meta-analysis of 21 trials with sex-disaggregated data, statins increased T2D risk similarly in women (OR 1.28) and men (OR 1.24). However, absence of study-level data precluded assessment of statin-specific or absolute risks, highlighting the need for more research in this area.
Women experience a disproportionate burden of cardiovascular–kidney–metabolic (CKM) disease, with higher mortality and greater life-years lost than men. Despite this, sex-specific evidence guiding CKM risk assessment and management remains limited. Female-specific factors, including pregnancy complications, polycystic ovary syndrome (PCOS), hormonal transitions, and menopause are inconsistently integrated into care, while women remain underrepresented in trials. This review highlights sex-based CKM differences and priorities for equitable, life-course–informed prevention and management in women.
Females with Barrett’s esophagus (BE) are thought to be less likely to develop neoplasia. This study assessed neoplastic progression risk in females to inform surveillance management. Patients were included from three prospective BE cohorts. Multivariable Cox regression models were used to assess risk factors (including sex) for neoplastic progression (i.e., the development of high-grade dysplasia or esophageal adenocarcinoma). 3099 patients were included, of whom 866 (28%) were females. Fifty-two females (6.0%) and 206 males (9.2%) developed neoplasia during a median follow-up of 6.2 years (IQR 3.3–10.9). Although progression occurred less frequently in females than in males (0.82% vs 1.20% per patient-year), sex was just weakly associated with progression risk in multivariable analysis (HR 1.38, 95% CI 1.00–1.89). Moreover, progression rates in females and males with a BE segment ≥3 cm were similar. Furthermore, mean time to progression, tumor stage, and treatment did not differ between sexes. Therefore, sex-specific surveillance cannot be recommended.
Approximately 1.8 billion people experience hormonal changes caused by the menstrual cycle, which impacts work, school, and health factors like physical activity. Prior research shows high-level insights, such as reduced activity during symptomatic days, but the day-to-day relationship between the menstrual cycle and physical activity remains unclear. We performed a 28-day study monitoring physical activity in 26 healthy, naturally menstruating women using an inertial measurement unit on the shank. A validated model estimated energy expended per step. Physical activity was significantly higher during the early-follicular phase than the late follicular phase due to more steps. Energy expenditure was compared with open-source daily estradiol and progesterone data collected from different study cohorts of healthy, eumenorrheic women. Peak estradiol and physical activity showed a correlation (r2 = 0.64) when activity was shifted by 2 days, suggesting a delayed relationship. Precisely measuring movement with wearable sensors could help uncover menstrual cycle effects on activity and improve health management and workplace policies.
Cervical cancer remains a leading cause of preventable mortality in low- and middle-income countries, where visual inspection with acetic acid (VIA) is often used for screening. VIA’s reliability is limited by the lack of magnification, absence of imaging for review, and high inter-operator variability. We developed the Callascope, a portable colposcope designed to address these limitations by integrating high-resolution imaging and acetic acid application in a single device. The Callascope can be used with or without a speculum and incorporates an asymmetrical tip for cervix positioning, an integrated camera with illumination, and a built-in atomizer for uniform contrast application. Bench testing confirmed image quality and contrast application were comparable to a standard colposcope. In clinical studies, the Callascope enabled consistent cervical visualization, effective acetowhitening, and reliable image interpretation while improving patient comfort. By addressing VIA’s limitations, it has the potential to expand screening access and support global cervical cancer elimination.
We aimed to develop and validate a tool to comprehensively measure person-centered antenatal care (PCANC) in low- and middle-income countries (LMICs). We followed standard procedures for scale development, including literature review, expert reviews, cognitive interviews, and pretesting to ensure content validity. Questions were iteratively revised at each stage and administered in surveys with pregnant and postpartum women in Ghana and Kenya. The survey data were used for psychometric analysis, resulting in a 36-item PCANC scale with three subscales: “dignity and respect,” “communication and autonomy,” and “responsive and supportive care.” The Cronbach’s alpha is 0.90 for the full scale and >0.7 for each subscale. The summative PCANC scores correlate with global measures of antenatal care quality, satisfaction, and future care location, suggesting good criterion validity. The PCANC scale has high validity and reliability and will facilitate efforts to measure and improve respectful and responsive antenatal care in LMICs.
Despite known hormonal influences on immune function, the impact of menstrual cycle phase on vaccine outcomes remains unexplored. Using prospectively tracked cycle data from the Clue period tracking-app, matched with an in-app survey of 1474 women, we compared self-reported COVID-19 vaccine outcomes between follicular-phase (estrogen-dominant) and luteal-phase (progesterone-dominant) vaccinated women. Side effect presence, severity and number were analysed using binary, ordinal, and negative binomial regression respectively, and post-vaccination time to infection (67–372 days) via Wilcoxon rank-sum test. Follicular-phase vaccination was associated with 35% higher odds of any self-reported side effects (OR: 1.35, 98.3% CI: 1.01–1.79) compared to luteal-phase vaccination. Median time to infection was 35 days longer following follicular-phase vaccination (200 [140–237] vs 165 [107–210] days, p = 0.05), though infection numbers were limited. These findings suggest menstrual cycle phase warrants consideration in sex-based immunity research and may inform future research on vaccination and personalised health strategies.
Endometriosis is a chronic inflammatory condition affecting 2-10% of reproductive-aged women, most commonly presenting with pelvic pain and subfertility. While its impact on reproductive health is increasingly recognized, the vaginal microbiome's role in the pathogenesis of endometriosis is still unclear and a comprehensive map of potential co-occurring conditions remains underexplored. Leveraging data from the Isala citizen-science platform in Flanders (Belgium), we analysed vaginal microbiome profiles obtained through 16S rRNA sequencing and health data from 95 women with self-reported endometriosis and 2,279 without. While no differences were observed in vaginal microbiome composition or diversity, we identified significant associations between endometriosis and polyendocrine metabolic ovarian syndrome (previously named polycystic ovarian syndrome; OR = 2.92, 95% CI 1.71-4.78, p < 0.001), migraine (OR = 3.75, 95% CI 1.38-8.60, p = 0.025), irritable bowel syndrome (OR = 2.57, 95% CI 1.43-4.36, p = 0.008) and dyspareunia (OR = 1.67, 95% CI 1.14-2.40, p = 0.033). These findings suggest that the vaginal microbiome composition plays at most a limited role in endometriosis and highlights how citizen science can effectively complement clinical research by capturing underrecognized comorbidities.
In this article, I claim that pronatalist policies that not only reward those who have (more) children, but penalize those who do not, constitute what I call negative pronatalism. I argue that negative pronatalism exacerbates autonomy infringements, inequalities, and other harms to those – especially women – who do not have (more) children. Thus, negative pronatalism warrants repudiation.
Despite its centrality to women’s health, the menstrual cycle remains understudied in computational health research due to its complexity, variability, and limited data availability. Recent advances in generative artificial intelligence (AI) offer new opportunities for modeling large-scale, user-generated menstrual health data. We introduce and evaluate a generative foundation model trained on self-tracked data from over 1.2 million users of a widely used menstrual tracking app. We assess the model’s ability to generate physiologically plausible synthetic cycles and realistic tracking behaviors, examine whether learned representations capture meaningful temporal and symptomatic patterns, and evaluate privacy risks. Results show that the model produces high-fidelity synthetic data closely mirroring real-world users, with no evidence of data leakage, while learned representations consistently outperform baseline methods on downstream forecasting tasks. These findings highlight generative AI’s potential to advance menstrual health forecasting, support privacy-sensitive data sharing, and enable scientific inquiry in women’s health research.
The combination therapy of FDA approved oxidative phosphorylation inhibitors (metformin and atovaquone) and platelet derived growth factor inhibitors (sunitinib and sorafenib) could target ovarian cancer stem-like cells (CSCs) and carcinoma associated mesenchymal stem cell enrichment of CSCs while reducing off-target effects. Results of a 48-hour drug exposure ovarian cancer-MSC tumoroid model showed additive effects with atovaquone combined with sunitinib or sorafenib being most effective.
Women with pre-pregnancy overweight or obesity are at increased risk of adverse pregnancy outcomes (APOs) and postpartum weight retention (PPWR). We examined which lipid classes were associated with APO and PPWR during pregnancy and postpartum using a subsample from a clinical trial. Data were collected via questionnaires, electronic health records, and participant-collected dried blood spots at three time-points. Lipidomic profiles were measured at all three time points in 49 participants. Using weighted-lipid correlation network analysis, differential lipid analysis, and partial-least squares discriminant analysis, we identified triglyceride (TG)-rich lipid signatures associated with APOs and PPWR. In early pregnancy, three TG networks and seven individual TGs were consistently associated with APOs. Postpartum, several TG networks and individual TGs were associated with APOs and PPWR. These findings highlight TG lipids' crucial role in pregnancy outcomes and the potential of TG-based lipidomic biomarkers for early risk identification to improve maternal and fetal health.
Contralateral prophylactic mastectomy (CPM) reduces contralateral breast cancer risk but improves survival only among young BRCA1/2 carriers. With expanded germline testing (GT) and increased use of oncologist-led and telehealth genetic services post-COVID, concerns remain about patients' understanding of results and the impact on surgical decisions. We conducted a retrospective cohort study of 1,054 women with unilateral breast cancer who underwent GT at Columbia University Irving Medical Center from 2013-2022. Mean age was 51.2; 40.5% were non-Hispanic White (NHW), 28.4% Hispanic, 12.1% Black, and 10.3% Asian. Overall, 28% underwent CPM. Testing shifted from 99.7% in-person to 73.9% oncologist-led and 20.1% telehealth post-COVID. In multivariable analysis, CPM was associated with younger age, advanced stage, and pathogenic/likely pathogenic variants, but not with service delivery or VUS results. A trend toward higher CPM rates was observed among Hispanic versus NHW patients (OR = 1.34, 95% CI = 0.99-1.81, p = 0.055). Despite care delivery shifts, CPM rates remained stable.
Subjective cognitive symptoms (brain fog, memory problems) are common in menopause, but whether these self-reported symptoms correspond to measurable deficits in cognition remains unclear. In 14,234 females (aged 45–55) from the REACT-Long Covid Study, we examined self-reported cognitive symptoms and objective cognitive performance in premenopausal, perimenopausal, postmenopausal participants. Global cognitive performance was derived from eight online tasks (‘Cognitron’). Perimenopausal (OR = 1.31 [1.18, 1.35], p = 0.015) and postmenopausal participants (OR = 1.13 [1.08, 1.32], p = 0.014) had higher odds of reporting cognitive symptoms than premenopausal participants. Objective cognitive performance differed minimally across menopause status groups, with perimenopausal participants showing marginally higher accuracy than premenopausal and postmenopausal participants (0.03–0.06 SD, p < 0.023). Across all menopause status groups, cognitive symptoms were weakly associated with objective performance but moderately related to psychological symptoms. These findings underscore recognising cognitive symptoms as a key component of menopausal care and integrating patient-reported outcomes with objective and biological measures of cognitive health.
The spectrum from normal brain functions to neurological and mental health disorders represents the life-course neural exposome. Proactive interventions strengthen rescue and reparative choices. Integration of intersectionality into health policy promotes global equity. Transdisciplinary care by women offers bio-social healthcare advantages for all persons. Reproduction and pregnancy establish foundational brain health during their child’s first 1000 days. Life-course healthcare that includes matrescence is more likely to promote brain health across successive lifespans.
Intimate partner violence (IPV) refers to the abuse from previous or current partners. It is a widespread but underreported public health concern that has a wide range of negative effects on the physical and mental health of those affected. This work presents machine learning models for the early detection of IPV in clinical settings, developed with a dataset of female patients who sought help at a domestic abuse intervention and prevention center of a major hospital in the United States. Utilizing tabular clinical data and unstructured clinical notes, we build single-modality and multimodal models for different data availability scenarios. Our multimodal model can identify patients at risk of IPV with an AUC of 0.88 and years before patients seek help. We validated the model on patients who did not seek help at the intervention center and patients from another hospital in the same integrated network with comparable performance.
Menstrual hygiene management (MHM) remains a critical yet often neglected issue in humanitarian settings. Reusable menstrual underwear (MU) offers a sustainable option where disposable products are limited, but evidence from low-income settings is scarce. This mixed-methods study assessed MU acceptability and usability among adult women in Kalehe, Democratic Republic of Congo. Participants received four MU, a hygiene kit, and instructions, with follow-up after three months through a survey (n = 124), and focus group discussions (n = 9). MU was highly accepted, with 94.3% reporting satisfaction and 98.4% preferring it to their usual MHM method. Reported benefits included comfort, hygiene, and ease of use, while challenges included absorbency, insufficient quantity, drying, and durability. Participants recommended improving MU quality, tailoring sizing, and expanding distribution, especially to adolescents. Finding suggests MU is a feasible, acceptable MHM option in a low-resource, conflict-affected setting, warranting further evaluation as part of emergency kits for displaced populations.
Endometriosis and fibroids are prevalent uterine conditions. Mechanisms for both conditions—immune dysfunction, inflammation, and sex hormone dysfunction—are plausibly influenced by diet. Here, we review the epidemiology and pathophysiology, and synthesize evidence on dietary patterns, endometriosis and uterine fibroid prevalence, risk, and symptomology. The evidence (12 studies) suggests that low-quality diets (ultra-processed) are associated with higher rates of both conditions, and high-quality diets are associated with lower symptom burden.
