Hasil untuk "Gynecology and obstetrics"

M. Hod, A. Kapur, D. Sacks et al.

In addition to the authors, t he following people provided important contributions during the creation of the document. Thanks go to international experts: Tao Duan, Huixia Yang, Andre Van Assche, Umberto Simeoni, Tahir Mahmood, Biodun Olagbuji, Eugene Sobngwi, Maicon Falavigna, Rodolfo Martinez, Carlos Ortega, Susana Salzberg, Jorge Alvariñas, Gloria Lopez Steward, Silvia Lapertosa, Roberto Estrade, Cristina Faingold, Silvia García, Argyro Syngelaki, Stephen Colagiuri, Yoel Toledano, Mark Hanson, and Blami Dao. Special thanks, for FIGO guidance and coordination, go to President Sabaratnam Arulkumaran, President Elect CN Purandare, Chief Executive Hamid Rushwan, and Chair of the SMNH Committee, William Stones. The following external groups evaluated the document and support its contents: European Board and College of Obstetrics and Gynaecology (EBCOG), The Society of Obstetricians and Gynaecologists of Canada (SOGC), Chinese Society of Perinatal Medicine, Diabetic Pregnancy Study Group (DPSG), African Federation of Obstetrics and Gynaecology (AFOG), South Asian Federation of Obstetrics and Gynecology (SAFOG), Australian Diabetes in Pregnancy Society (ADIPS), International Association of Diabetes in Pregnancy Study Groups (IADPSG), European Association of Perinatal Medicine (EAPM), Diabetes in Pregnancy Study Group of India (DIPSI), and the Diabetes in Pregnancy Study Group of Latin America. In addition to the FIGO Executive Board, all relevant FIGO Committees and Working Groups contributed to and supported the document. Acknowledgments

S. Li, M. Kemp, Susan J. Logan et al.

BACKGROUND Natural language processing is a form of artificial intelligence that allows human users to interface with a machine without using complex code. The ability of natural language processing systems such as ChatGPT to successfully engage with healthcare systems requiring fluid reasoning, specialist data interpretation, and empathetic communication in an unfamiliar and evolving environment is poorly studied. We tested whether the ChatGPT interface could engage with and complete a mock Objective Structured Clinical Examination simulating assessment for Membership of the Royal College of Obstetricians and Gynaecologists. OBJECTIVES We hypothesized that, without additional training, ChatGPT would achieve a score at least equivalent to that achieved by human candidates who sat for a virtual Objective Structured Clinical Examinations in Singapore. STUDY DESIGN The study was conducted in two phases. In the first, a total of seven structured discussion questions were selected from two historical cohorts (Cohorts A and B) of objective structured clinical examination questions. ChatGPT was examined using these questions and responses recorded in a script. Two human candidates (acting as anonymizers) were then examined on the same questions using videoconferencing, and their responses were transcribed verbatim into written scripts. The three sets of response scripts were then mixed, and each set allocated to one of three human actors. In the second stage, actors were used to present these scripts to examiners in response to the same examination questions. These responses were blind-scored by fourteen qualified examiners. ChatGPT scores were then unblinded and compared to historical human candidate performances. RESULTS The average ChatGPT score given to ChatGPT by 14 examiners was 77.2 %. The average historical human score (n=26 candidates) was 73.7 %. ChatGPT demonstrated sizable performance improvements over the average human candidate in several subject domains. The median time taken for ChatGPT to complete each station was 2.54 minutes, well before the ten-minutes allowed. CONCLUSIONS ChatGPT generated factually accsturate and contextually relevant structured discussion answers to complex and evolving clinical questions based on unfamiliar settings within a very short period. ChatGPT outperformed human candidates in several knowledge areas. Not all examiners were able to discern between human and ChatGPT responses. Our data highlight the emergent ability of natural language processing models to demonstrate fluid reasoning in unfamiliar environments and successfully compete with human candidates that have undergone extensive specialist training.

Xianying Cheng, Ping Wang, Li Cheng et al.

BackgroundCervical cancer incidence and mortality rates in the United States have substantially declined over recent decades, primarily driven by reductions in squamous cell carcinoma cases. However, the trend in recent years remains unclear. This study aimed to explore the trends in cervical cancer incidence and mortality, stratified by demographic and tumor characteristics from 1975 to 2018.MethodsThe age-adjusted incidence, incidence-based mortality, and relative survival of cervical cancer were calculated using the Surveillance, Epidemiology, and End Results (SEER)-9 database. Trend analyses with annual percent change (APC) and average annual percent change (AAPC) calculations were performed using Joinpoint Regression Software (Version 4.9.1.0, National Cancer Institute).ResultsDuring 1975–2018, 49,658 cervical cancer cases were diagnosed, with 17,099 recorded deaths occurring between 1995 and 2018. Squamous cell carcinoma was the most common histological type, with 34,169 cases and 11,859 deaths. Over the study period, the cervical cancer incidence rate decreased by an average of 1.9% (95% CI: −2.3% to −1.6%) per year, with the APCs decreased in recent years (−0.5% [95% CI: −1.1 to 0.1%] in 2006–2018). Squamous cell carcinoma incidence trends closely paralleled overall cervical cancer patterns, but the incidence of squamous cell carcinoma in the distant stage increased significantly (1.1% [95% CI: 0.4 to 1.8%] in 1990–2018). From 1995 to 2018, the overall cervical cancer mortality rate decreased by 1.0% (95% CI: −1.2% to −0.8%) per year. But for distant-stage squamous cell carcinoma, the mortality rate increased by 1.2% (95% CI: 0.3 to 2.1%) per year.ConclusionFor cervical cancer cases diagnosed in the United States from 1975 to 2018, the overall incidence and mortality rates decreased significantly. However, there was an increase in the incidence and mortality of advanced-stage squamous cell carcinoma. These epidemiological patterns offer critical insights for refining cervical cancer screening protocols and developing targeted interventions for advanced-stage cases.

Xuejie Gao, Xuejie Gao, Yuyun Chen et al.

ObjectiveThis study examines the association between serum vitamin D levels and the prevalence of metabolic dysfunction-associated fatty liver disease (MAFLD) in adolescents, along with potential modifying factors.MethodsData from 950 adolescents aged 12–18 years in the National Health and Nutrition Examination Survey (NHANES) 2017–2018 were analyzed. MAFLD was defined using hepatic steatosis and metabolic dysfunction criteria. Serum 25(OH)D levels were measured, and weighted logistic regression and restricted cubic spline models were applied to assess their association with MAFLD risk. Stratified analyses were also conducted.ResultsLower serum 25(OH)D levels were significantly associated with higher MAFLD risk (p < 0.001), showing a nonlinear dose-response relationship. Adolescents with 25(OH)D ≥ 75 nmol/L had a 57% lower risk of MAFLD compared to those with levels < 50 nmol/L. Stratified analysis indicated that the protective effect of vitamin D was more evident in individuals with higher retinol levels, though retinol alone was not significantly associated with MAFLD.ConclusionVitamin D deficiency is significantly associated with MAFLD in adolescents, with a nonlinear dose-response relationship modulated by retinol status. These findings underscore the potential role of vitamin D in MAFLD prevention and provide a basis for further prospective or intervention studies.

Hongqin Chen, Jian Meng, Qiao Li et al.

Abstract Background The heterogeneity of patient-reported outcomes (PROs) and patient-reported outcome measures (PROMs) in published clinical studies on vaginal relaxation syndrome (VRS) hinders cross-study comparisons and integration of evidence-based findings, impeding the development of robust clinical evidence. Objective To comprehensively investigate the current use of PROs and PROMs in VRS research, compile a comprehensive catalog, and provide guidance for selecting outcome measures and tools VRS patients. Methods This study systematically searched clinical studies on VRS treatment published up to December 2024 in PUBMED, EMBASE, Web of Science, and Cochrane databases, focusing primarily on pelvic floor muscle training, physical energy therapies, and surgical interventions. PROs and PROMs were extracted, organized into a structured catalog, and categorized by thematic domains. The COSMIN checklist was applied to assess the measurement properties of commonly used PROMs. Results A total of 69 studies were included, comprising 14 randomized controlled trials (1193 patients) and 55 observational studies (3327 patients), totaling 4520 participants. These studies reported 68 PROs and 57 PROMs. The most commonly used PROMs were the Female Sexual Function Index (FSFI, 47.83%), Vaginal Laxity Questionnaire (VLQ), Visual Analog Scale (VAS), Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire (PISQ-12), and Sexual Satisfaction Questionnaire (SSQ). Notably, 42 PROMs (73.68%) appeared only once. Conclusions PROs for surgical and non-surgical VRS treatments are similar, but non-surgical interventions include additional outcomes, such as overall efficacy and patient’s vaginal tightness satisfaction. The high proportion of unvalidated PROMs (81.09%) underscores the need for standardized, disease-specific measures. Future Delphi surveys and expert consensus are anticipated to facilitate the development of a comprehensive core outcome set (COS) and core outcome measurement set (COMS) for VRS.

Dou Liu, Ying Long, Sophia Zuoqiu et al.

Effective physician-patient communications in pre-diagnostic environments, and most specifically in complex and sensitive medical areas such as infertility, are critical but consume a lot of time and, therefore, cause clinic workflows to become inefficient. Recent advancements in Large Language Models (LLMs) offer a potential solution for automating conversational medical history-taking and improving diagnostic accuracy. This study evaluates the feasibility and performance of LLMs in those tasks for infertility cases. An AI-driven conversational system was developed to simulate physician-patient interactions with ChatGPT-4o and ChatGPT-4o-mini. A total of 70 real-world infertility cases were processed, generating 420 diagnostic histories. Model performance was assessed using F1 score, Differential Diagnosis (DDs) Accuracy, and Accuracy of Infertility Type Judgment (ITJ). ChatGPT-4o-mini outperformed ChatGPT-4o in information extraction accuracy (F1 score: 0.9258 vs. 0.9029, p = 0.045, d = 0.244) and demonstrated higher completeness in medical history-taking (97.58% vs. 77.11%), suggesting that ChatGPT-4o-mini is more effective in extracting detailed patient information, which is critical for improving diagnostic accuracy. In contrast, ChatGPT-4o performed slightly better in differential diagnosis accuracy (2.0524 vs. 2.0048, p > 0.05). ITJ accuracy was higher in ChatGPT-4o-mini (0.6476 vs. 0.5905) but with lower consistency (Cronbach's $α$ = 0.562), suggesting variability in classification reliability. Both models demonstrated strong feasibility in automating infertility history-taking, with ChatGPT-4o-mini excelling in completeness and extraction accuracy. In future studies, expert validation for accuracy and dependability in a clinical setting, AI model fine-tuning, and larger datasets with a mix of cases of infertility have to be prioritized.

V. Peppa, M. Robitaille, F. Akbari et al.

Purpose: To develop clinically relevant test cases for commissioning Model-Based Dose Calculation Algorithms (MBDCAs) for 192Ir High Dose Rate (HDR) gynecologic brachytherapy following the workflow proposed by the TG-186 report and the WGDCAB report 372. Acquisition and Validation Methods: Two cervical cancer intracavitary HDR brachytherapy patient models were created, using either uniformly structured regions or realistic segmentation. The computed tomography (CT) images of the models were converted to DICOM CT images via MATLAB and imported into two Treatment Planning Systems (TPSs) with MBDCA capability. The clinical segmentation was expanded to include additional organs at risk. The actual clinical treatment plan was generally maintained, with the source replaced by a generic 192Ir HDR source. Dose to medium in medium calculations were performed using the MBDCA option of each TPS, and three different Monte Carlo (MC) simulation codes. MC results agreed within statistical uncertainty, while comparisons between MBDCA and MC dose distributions highlighted both strengths and limitations of the studied MBDCAs, suggesting potential approaches to overcome the challenges. Data Format and Usage Notes: The datasets for the developed cases are available online at http://doi.org/ 10.5281/zenodo.15720996. The DICOM files include the treatment plan for each case, TPS, and the corresponding reference MC dose data. The package also contains a TPS- and case-specific user guide for commissioning the MBDCAs, and files needed to replicate the MC simulations. Potential Applications: The provided datasets and proposed methodology offer a commissioning framework for TPSs using MBDCAs, and serve as a benchmark for brachytherapy researchers using MC methods. They also facilitate intercomparisons of MBDCA performance and provide a quality assurance resource for evaluating future TPS software updates.

Erin Palm, Astrit Manikantan, Mark E. Pepin et al.

In medical practices across the United States, physicians have begun implementing generative artificial intelligence (AI) tools to perform the function of scribes in order to reduce the burden of documenting clinical encounters. Despite their widespread use, no established methods exist to gauge the quality of AI scribes. To address this gap, we developed a blinded study comparing the relative performance of large language model (LLM) generated clinical notes with those from field experts based on audio-recorded clinical encounters. Quantitative metrics from the Physician Documentation Quality Instrument (PDQI9) provided a framework to measure note quality, which we adapted to assess relative performance of AI generated notes. Clinical experts spanning 5 medical specialties used the PDQI9 tool to evaluate specialist-drafted Gold notes and LLM authored Ambient notes. Two evaluators from each specialty scored notes drafted from a total of 97 patient visits. We found uniformly high inter rater agreement (RWG greater than 0.7) between evaluators in general medicine, orthopedics, and obstetrics and gynecology, and moderate (RWG 0.5 to 0.7) to high inter rater agreement in pediatrics and cardiology. We found a modest yet significant difference in the overall note quality, wherein Gold notes achieved a score of 4.25 out of 5 and Ambient notes scored 4.20 out of 5 (p = 0.04). Our findings support the use of the PDQI9 instrument as a practical method to gauge the quality of LLM authored notes, as compared to human-authored notes.

Jayroop Ramesh, Valentin Bacher, Mark C. Eid et al.

The International Society of Ultrasound advocates Intrapartum Ultrasound (US) Imaging in Obstetrics and Gynecology (ISUOG) to monitor labour progression through changes in fetal head position. Two reliable ultrasound-derived parameters that are used to predict outcomes of instrumental vaginal delivery are the angle of progression (AoP) and head-symphysis distance (HSD). In this work, as part of the Intrapartum Ultrasounds Grand Challenge (IUGC) 2024, we propose an automated fetal biometry measurement pipeline to reduce intra- and inter-observer variability and improve measurement reliability. Our pipeline consists of three key tasks: (i) classification of standard planes (SP) from US videos, (ii) segmentation of fetal head and pubic symphysis from the detected SPs, and (iii) computation of the AoP and HSD from the segmented regions. We perform sparse sampling to mitigate class imbalances and reduce spurious correlations in task (i), and utilize ensemble-based deep learning methods for task (i) and (ii) to enhance generalizability under different US acquisition settings. Finally, to promote robustness in task iii) with respect to the structural fidelity of measurements, we retain the largest connected components and apply ellipse fitting to the segmentations. Our solution achieved ACC: 0.9452, F1: 0.9225, AUC: 0.983, MCC: 0.8361, DSC: 0.918, HD: 19.73, ASD: 5.71, $Δ_{AoP}$: 8.90 and $Δ_{HSD}$: 14.35 across an unseen hold-out set of 4 patients and 224 US frames. The results from the proposed automated pipeline can improve the understanding of labour arrest causes and guide the development of clinical risk stratification tools for efficient and effective prenatal care.

Victoria-Elisabeth Gruber, Razvan Marinescu, Diego Fajardo et al.

As large language models (LLMs) become primary sources of health information for millions, their accuracy in women's health remains critically unexamined. We introduce the Women's Health Benchmark (WHB), the first benchmark evaluating LLM performance specifically in women's health. Our benchmark comprises 96 rigorously validated model stumps covering five medical specialties (obstetrics and gynecology, emergency medicine, primary care, oncology, and neurology), three query types (patient query, clinician query, and evidence/policy query), and eight error types (dosage/medication errors, missing critical information, outdated guidelines/treatment recommendations, incorrect treatment advice, incorrect factual information, missing/incorrect differential diagnosis, missed urgency, and inappropriate recommendations). We evaluated 13 state-of-the-art LLMs and revealed alarming gaps: current models show approximately 60\% failure rates on the women's health benchmark, with performance varying dramatically across specialties and error types. Notably, models universally struggle with "missed urgency" indicators, while newer models like GPT-5 show significant improvements in avoiding inappropriate recommendations. Our findings underscore that AI chatbots are not yet fully able of providing reliable advice in women's health.

M. W. Hanson, Anne Bardsley, L. De-Regil et al.

a Institute of Developmental Sciences, University of Southampton; and NIHR Nutrition Biomedical Research Centre, University Hospital Southampton; Southampton, UK b Liggins Institute, University of Auckland, Auckland, New Zealand c Micronutrient Initiative, Ottawa, Canada d MRC Human Nutrition Research, Cambridge, UK e Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute; and Department of Nutrition, Harvard TH Chan School of Public Health; Boston, MA, USA f King’s College London, London, UK g Department of Medicine and Therapeutics, The Chinese University of Hong Kong; and the Hong Kong Institute of Diabetes and Obesity, The Chinese University of Hong Kong, Hong Kong, China h UCD School of Medicine and Medical Science, University College Dublin, National Maternity Hospital, Dublin, Ireland i University of Malawi College of Medicine, Blantyre, Malawi j Grant Medical College, Mumbai, India k Indian College of Obstetricians and Gynaecologists, Mumbai, India l International Federation of Gynecology and Obstetrics, London, UK

Kavita Vinekar, Aishwarya Karlapudi, L. Nathan et al.

If Roe v Wade is overturned, at least 43.9% of U.S. obstetrics and gynecology residents are predicted to lack abortion training. In June 2022, the U.S. Supreme Court is expected to issue a decision on Dobbs v Jackson Women's Health Organization, a direct challenge to Roe v Wade. A detailed policy analysis by the Guttmacher Institute projects that, if Roe v Wade is overturned, 21 states are certain to ban abortion and five states are likely to ban abortion. The Accreditation Council for Graduate Medical Education requires access to abortion training for all obstetrics and gynecology residency programs. We performed a comprehensive study of all accredited U.S. obstetrics and gynecology residency programs to assess how many of these programs and trainees are currently located in states projected to ban abortion if Roe v Wade is overturned. We found that, of 286 accredited obstetrics and gynecology residency programs with current residents, 128 (44.8%) are in states certain or likely to ban abortion if Roe v Wade is overturned. Therefore, of 6,007 current obstetrics and gynecology residents, 2,638 (43.9%) are certain or likely to lack access to in-state abortion training. Preparation for the reversal of Roe v Wade should include not only a recognition of the negative effects on patient access to abortion care in affected states, but also of the dramatic implications for obstetrics and gynecology residency training.

Uppala Radhakrishna, Uppala Ratnamala, Devendrasinh D. Jhala et al.

BackgroundHidradenitis suppurativa (HS) is a chronic inflammatory skin condition that affects hair follicles in areas with apocrine sweat glands, such as the underarms, groin, and buttocks. The pathogenesis of HS is not fully understood, but considering the key role played by the biological clock in the control of immune/inflammatory processes the derangement of circadian and ultradian pathways could be hypothesized.MethodsWe analyzed genome-wide DNA methylation patterns in peripheral blood from 24 HS cases and 24 controls using the Infinium HumanMethylation450 BeadChip array (Illumina), followed by bioinformatics and statistical analyses.ResultsWe found that several circadian and ultradian genes were differentially methylated in HS patients, predominantly exhibiting hypomethylation. These genes were enriched in pathways such as MAPK and WNT cascades, acute phase response, cytokine release, inflammation, innate immune response, xenobiotic detoxification, and oxidative stress response.ConclusionAltered DNA methylation patterns of genes related to circadian and ultradian pathways could contribute to immune system derangement and inflammatory processes chronicization in addition to other comorbidities hallmarking HS onset and progression, at the same time representing possible druggable targets.

Dilek Konukbay, Emine Öksüz, Gulten Guvenc

Abstract Background Breastfeeding self-efficacy is one of the key factors that affect a healthy and successful breastfeeding process. A mother’s belief regarding her ability to breastfeed is influenced by social and psychological factors. This study aimed to investigate the breastfeeding self-efficacy levels of postpartum women, the factors affecting this, and its relationship with sleep quality, social support and depression. Methods This descriptive cross-sectional study was conducted in the pediatric department of a tertiary hospital in Ankara, Turkey. Data were collected from 200 postpartum women using the Breastfeeding Self-Efficacy Scale-Short Form (BSES-SF), the Pittsburgh Sleep Quality Index (PSQI), the Multidimensional Scale of Perceived Social Support (MSPSS) and the Edinburgh Postnatal Depression Scale (EPDS). Results The mean scores of the BSES-SF, PSQI, MSPSS and EPDS were 59.05 ± 8.28, 9.18 ± 3.67, 57.82 ± 18.81, and 8.98 ± 5.89, respectively. A statistically significant negative correlation was found among the BSES-SF, EPDS (r = -0.445, p = 0.001) and PSQI (r = -0.612, p = 0.004), while a positive correlation was found among the BSES-SF, total MSPSS (r  = 0.341, p = 0.036), and family support (r  = 0.373, p = 0.014) (p < 0.05). In addition, a statistically significant difference was found between the number of births and breastfeeding self-efficacy (F = 3.68; p = 0.001). The linear regression analysis revealed that sleep quality (β = -0.491, p = 0.001), perceived social support (β = 0.146, p = 0.015), family support (β = 0.153, p = 0.013), and depression (β = -0.228, p = 0.001) emerged as the predictors of breastfeeding self-efficacy. Conclusions In this study, the increase in sleep quality and perceived social support positively affected the breastfeeding self-efficacy of postpartum women, while giving birth for the first time and an increase in the risk of depression were negatively affected.

Chunge Cao, Dajun Cai, Hao Liu et al.

IntroductionThe relationship between serum uric acid (SUA) and cervical cancer is inconclusive. This study aims to investigate the causal relationship between SUA levels and cervical cancer incidence, and to evaluate the potential role of nutritional interventions in cervical cancer prevention.MethodsWe conducted a two-sample bidirectional Mendelian randomization (MR) analysis using genetic instruments from publicly available genome-wide association studies (GWASs) of individuals of predominantly European ancestry. Methods such as inversevariance weighted, weighted-median, weighted model, and MR-Egger were applied. Sensitivity tests, including leave-one-out, MR-PRESSO, and Cochran’s Q test, assessed heterogeneity and pleiotropy.ResultsOur findings revealed that a high SUA concentration significantly increased the risk of malignant cervical cancer: a 1 mg/mL increase in SUA was associated with a 71% higher risk (OR = 1.71, 95% CI = 1.10–2.67; p = 0.018). Stratification by histological type showed a significant causal effect on cervical adenocarcinoma risk (OR = 2.56, 95% CI = 1.14–5.73; p = 0.023). However, no clear evidence was found for a causal effect of cervical cancer on SUA levels.ConclusionThis study identified a causal relationship between elevated SUA levels and the risk of malignant cervical cancer, particularly cervical adenocarcinoma. These findings provide novel insights into the mechanisms of cervical carcinogenesis and suggest that managing SUA levels could be a potential strategy for cervical cancer prevention through dietary management.

Ruize Shi, Hong Huang, Wei Zhou et al.

The rapid development of large language models (LLMs) has transformed many industries, including healthcare. However, previous medical LLMs have largely focused on leveraging general medical knowledge to provide responses, without accounting for patient variability and lacking true personalization at the individual level. To address this, we propose a novel method called personalized medical language model (PMLM), which explores and optimizes personalized LLMs through recommendation systems and reinforcement learning (RL). Specifically, by utilizing self-informed and peer-informed personalization, PMLM captures changes in behaviors and preferences to design initial personalized prompts tailored to individual needs. We further refine these initial personalized prompts through RL, ultimately enhancing the precision of LLM guidance. Notably, the personalized prompt are hard prompt, which grants PMLM high adaptability and reusability, allowing it to directly leverage high-quality proprietary LLMs. We evaluate PMLM using real-world obstetrics and gynecology data, and the experimental results demonstrate that PMLM achieves personalized responses, and it provides more refined and individualized services, offering a potential way for personalized medical LLMs.

Jieyun Bai, Zihao Zhou, Zhanhong Ou et al.

Segmentation of the fetal and maternal structures, particularly intrapartum ultrasound imaging as advocated by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) for monitoring labor progression, is a crucial first step for quantitative diagnosis and clinical decision-making. This requires specialized analysis by obstetrics professionals, in a task that i) is highly time- and cost-consuming and ii) often yields inconsistent results. The utility of automatic segmentation algorithms for biometry has been proven, though existing results remain suboptimal. To push forward advancements in this area, the Grand Challenge on Pubic Symphysis-Fetal Head Segmentation (PSFHS) was held alongside the 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023). This challenge aimed to enhance the development of automatic segmentation algorithms at an international scale, providing the largest dataset to date with 5,101 intrapartum ultrasound images collected from two ultrasound machines across three hospitals from two institutions. The scientific community's enthusiastic participation led to the selection of the top 8 out of 179 entries from 193 registrants in the initial phase to proceed to the competition's second stage. These algorithms have elevated the state-of-the-art in automatic PSFHS from intrapartum ultrasound images. A thorough analysis of the results pinpointed ongoing challenges in the field and outlined recommendations for future work. The top solutions and the complete dataset remain publicly available, fostering further advancements in automatic segmentation and biometry for intrapartum ultrasound imaging.

Abdennour Boulesnane, Boutheina Mokhtari, Oumnia Rana Segueni et al.

Medical imaging has significantly revolutionized medical diagnostics and treatment planning, progressing from early X-ray usage to sophisticated methods like MRIs, CT scans, and ultrasounds. This paper investigates the use of deep learning for medical image captioning, with a particular focus on uterine ultrasound images. These images are vital in obstetrics and gynecology for diagnosing and monitoring various conditions across different age groups. However, their interpretation is often challenging due to their complexity and variability. To address this, a deep learning-based medical image captioning system was developed, integrating Convolutional Neural Networks with a Bidirectional Gated Recurrent Unit network. This hybrid model processes both image and text features to generate descriptive captions for uterine ultrasound images. Our experimental results demonstrate the effectiveness of this approach over baseline methods, with the proposed model achieving superior performance in generating accurate and informative captions, as indicated by higher BLEU and ROUGE scores. By enhancing the interpretation of uterine ultrasound images, our research aims to assist medical professionals in making timely and accurate diagnoses, ultimately contributing to improved patient care.

Xi Chen, Yang Xu, MingKe You et al.

This study evaluates whether large language models (LLMs) exhibit biases towards medical professionals. Fictitious candidate resumes were created to control for identity factors while maintaining consistent qualifications. Three LLMs (GPT-4, Claude-3-haiku, and Mistral-Large) were tested using a standardized prompt to evaluate resumes for specific residency programs. Explicit bias was tested by changing gender and race information, while implicit bias was tested by changing names while hiding race and gender. Physician data from the Association of American Medical Colleges was used to compare with real-world demographics. 900,000 resumes were evaluated. All LLMs exhibited significant gender and racial biases across medical specialties. Gender preferences varied, favoring male candidates in surgery and orthopedics, while preferring females in dermatology, family medicine, obstetrics and gynecology, pediatrics, and psychiatry. Claude-3 and Mistral-Large generally favored Asian candidates, while GPT-4 preferred Black and Hispanic candidates in several specialties. Tests revealed strong preferences towards Hispanic females and Asian males in various specialties. Compared to real-world data, LLMs consistently chose higher proportions of female and underrepresented racial candidates than their actual representation in the medical workforce. GPT-4, Claude-3, and Mistral-Large showed significant gender and racial biases when evaluating medical professionals for residency selection. These findings highlight the potential for LLMs to perpetuate biases and compromise healthcare workforce diversity if used without proper bias mitigation strategies.

Yukiko Katagiri, S. Jwa, Akira Kuwahara et al.

The Japan Society of Obstetrics and Gynecology (JSOG) registry gathers comprehensive data from registered assisted reproductive technology (ART) facilities in Japan. Herein, we report 2021 ART cycle characteristics and outcomes.