Objective: To evaluate the quality of ChatGPT-generated responses to commonly asked questions during pregnancy, based on expert assessments using predefined criteria: accuracy, completeness, and safety. Methods: A total of 20 board-certified obstetricians evaluated 15 ChatGPT-generated responses to frequently encountered pregnancy-related questions. Each response was assessed using a 5-point Likert scale across three domains. Statistical comparisons were conducted to evaluate differences among criteria and question categories. Results: ChatGPT received an overall mean score of 4.1 across all criteria. Accuracy was the highest-rated criterion (mean 4.27 ± 0.31), followed by completeness (3.85 ± 0.30) and safety (3.78 ± 0.36) (P = 0.019). Responses to general knowledge questions scored significantly higher than those related to symptoms or follow-up guidance (P = 0.041). The most favorably rated response pertained to sleep positions during pregnancy (mean 4.5), while painkiller safety scored the lowest (mean 3.5). Conclusion: ChatGPT demonstrates strong potential in delivering accurate and comprehensible pregnancy-related information. However, its limitations in clinical safety and completeness—particularly in symptom-related topics—suggest that it should be used as an adjunct to, not a replacement for, professional medical guidance. Further validation across diverse clinical scenarios and standardized evaluation tools is necessary to ensure safe integration into patient education.

Gestational diabetes mellitus (GDM) is associated with substantial maternal and neonatal morbidity. Early identification of women at risk remains a clinica

Objective: This study aimed to evaluate the content, characteristics, and quality of the most-viewed YouTube videos on vaginitis, assessing their accuracy and potential role in patient education.Materials and Methods: On December 26, 2025, a new YouTube account was used to search “vaginitis” and “vaginal infection.” The first 100 videos were screened; 53 were excluded due to non-English language, lack of audiovisual content, or duplication, leaving 47 videos for analysis. Videos were categorized by content (disease explanation, treatment, personal experiences, other) and by uploader (academic, physician, paramedic, commercial, patient), further grouped as medical vs. non-medical. Video metrics (views, likes, like-to-view ratio, views per 1,000, Video Power Index [VPI]) and quality scores were recorded. Data were analyzed using the Mann–Whitney U and Fisher’s exact tests; interrater reliability was assessed with ICC.Results: Of the videos, 72.3% covered disease explanations, 12.8% other topics, 10.6% personal experiences, and 4.3% preventive measures. Uploaders were physicians (46.8%), academics (27.7%), commercial sources (12.8%), patients (8.5%), and paramedics (4.3%). Median views were 55,337 (range: 581–11,041,478), median likes were 402 (range: 11–295,000). VPI did not differ significantly between professional and non-professional sources (6.06 ± 30.80 vs. 2.29 ± 4.53; p=0.451), but quality scores were higher for professional sources (9.70 ± 4.03 vs. 6.20 ± 2.39; p=0.014). Videos uploaded in 2021 or later had higher VPI than earlier ones (0.556 vs. 0.114; p=0.003), with no significant difference in quality (p=0.083).Conclusion: Most YouTube videos on vaginitis focus on definitions, with limited guidance on prevention or self-care. While professional sources provide higher-quality content, engagement metrics do not correlate with reliability. Promotion of expert-led videos and automated quality control measures is warranted.Keywords: Vaginitis, Vaginal infection, YouTube,

With increasing life expectancy, a growing proportion of patients undergoing surgery for gynecologic cancers are older adults, underscoring the need for reliable predictors of postoperative recovery and patient engagement. Cognitive function and physical frailty are recognized determinants of surgical outcomes, yet their relative impact on patient centered outcomes remains insufficiently explored. This prospective observational study included 68 women aged 65 years and older who underwent abdominal surgery for gynecologic malignancies. Preoperative cognitive function was assessed using the Montreal Cognitive Assessment, and physical frailty was evaluated with the Clinical Frailty Scale. Postoperative outcomes included early recovery parameters, complications, surgical decision satisfaction, and home-based adherence. Higher cognitive scores were associated with earlier mobilization, shorter hospital stay, better postoperative adherence, and greater decision satisfaction, whereas higher frailty scores were associated with delayed recovery and increased complication risk. In regression analyses, preoperative cognitive function was significantly associated with both postoperative adherence and surgical decision satisfaction, whereas physical frailty was not. These findings indicate that preoperative cognitive screening may have predictive value for patient centered recovery behaviors and decision satisfaction in this setting; however, the prediction estimates should be considered exploratory and warrant validation in larger, multicenter cohorts.

Aim: The aim of this study was to compare the perioperative and postoperative clinical outcomes of transvaginal natural orifice transluminal endoscopic surgery (V-NOTES) and total laparoscopic hysterectomy (TLH) in patients undergoing surgery for benign gynecological indications.Materials and Methods: This retrospective cohort study included patients who underwent hysterectomy at a tertiary referral center between October 1, 2024, and October 1, 2025. Patients with histopathologically confirmed malignancy, emergency procedures, and incomplete data were excluded. A total of 711 patients were analyzed, including 38 in the V-NOTES group and 673 in the TLH group. Demographic, perioperative, and postoperative data were collected and compared between groups.Results: There were no significant differences between the groups in baseline characteristics. Operative time was significantly longer in the V-NOTES group (101.47±17.89 vs. 82.46±12.86 minutes, p0.05). Postoperative pain scores were significantly lower in the V-NOTES group at both the 6th and 24th hours (p=0.003 and p=0.012, respectively). Intraoperative complication rates were similar (p=0.150). Postoperative complications requiring hospitalization were more frequent in the V-NOTES group (7.9% vs. 2.1%), but the difference was not statistically significant (p=0.057).Conclusion: V-NOTES is a safe and feasible alternative to TLH, offering the advantage of reduced postoperative pain with comparable perioperative outcomes. Although operative time was longer, this may be related to the learning curve. Further prospective studies are needed to confirm these findings.