A Decade of Evolution in Ultrasound in Obstetrics and Gynecology Literature (2015–2025): A Bibliometric and Visualization Analysis From a Leading Specialty Journal Perspective

To editor, Obstetric and gynecologic ultrasound remains fundamental to maternal-fetal medicine, as these tools support prenatal screening, structural anomaly assessment, fetal growth surveillance, hemodynamic evaluation, and perinatal risk stratification. Over the past decade, continuous advances in imaging technology—combined with increasing integration of artificial intelligence (AI), machine learning, and biomarker-based risk prediction—have reshaped research priorities in the field. As publication volumes expand, clinicians and researchers find it increasingly difficult to rapidly identify major thematic shifts and emerging hotspots. Bibliometric analysis offers a quantitative approach for mapping scientific output, collaboration patterns, and topic evolution through visual and statistical methods.1,2 In this correspondence, we present a bibliometric overview of articles published in Ultrasound in Obstetrics & Gynecology (UOG) from January 1, 2015, and August 27, 2025. This journal was chosen owing to its prominence in the field and its consistent publication of high-impact research that informs clinical practice. The ten-year period from 2015 to 2025 was selected to capture significant developments in imaging technology, particularly the integration of AI, machine learning, and biomarker-based risk models, which have influenced clinical applications in maternal-fetal medicine. This analysis specifically reflects the thematic evolution within UOG rather than attempting to represent the entirety of obstetric and gynecologic ultrasound research. By concentrating on this leading specialty journal, we aim to provide a focused perspective on the primary trends and advancements that shape clinical practice. Materials and methods The Web of Science Core Collection was searched using the source-title strategy: SO = (ULTRASOUND IN OBSTETRICS GYNECOLOGY). The final search was conducted on August 27, 2025. Eligible records met the following criteria: English-language articles published between January 1, 2015, and August 27, 2025. Duplicate records and records with incomplete essential bibliographic information were excluded. All retrieved records were exported as “Full Record and Cited References,” including title, author, abstract, keywords, year, and cited references. Two physicians independently screened the records and verified eligibility. Disagreements were resolved through discussion with the corresponding author. A standard PRISMA flowchart is provided in Supplementary Fig. 1, https://links.lww.com/MFM/A133. Of a total of 1662 articles initially exported, 86 were excluded due to missing essential fields, leaving 1576 articles for the final analysis. Keyword normalization and cleaning, including the standardization of synonyms, abbreviations, singular/plural forms, and spelling variants, were performed using COOC 14.9, followed by keyword analysis. Bibliometric analyses were performed using CiteSpace 6.2R6, VOSviewer, COOC 14.9, and Excel. The main indicators included annual publication output, country/institution/author productivity, collaboration networks, keyword co-occurrence, clustering, time-zone evolution, and burst-term detection (Supplementary Fig. 1, https://links.lww.com/MFM/A133). Results Publication trends The cumulative number of publications in UOG followed a steady upward trajectory throughout the study period. Annual output remained relatively stable, ranging from 135 to 173 articles per year, with prominent peaks in 2016 and 2021. In 2016, a total of 168 articles were published, coinciding with intensified interest in first-trimester aneuploidy screening and biomarker-driven prediction of hypertensive disorders of pregnancy.3,4 Similarly, the increase to 173 articles in 2021 resulted partly from the surge in COVID-19-related maternal-fetal research. The apparent decrease to 135 articles in 2022 does not indicate a decline in the relevance of the field; rather, it reflects natural variations in research focus, potentially influenced by shifting global priorities and external factors such as public health events and technological advancements. Such fluctuations are common in scientific disciplines and represent the cyclical nature of research momentum rather than a reduction in overall interest (Supplementary Fig. 2A, https://links.lww.com/MFM/A133). Geographic and institutional distribution The United Kingdom contributed the largest number of publications (543), followed by the United States (271), Italy (239), Spain (187), the Netherlands (172), Australia (147), Belgium (135), and China (125). While Europe, particularly the UK, leads in maternal-fetal medicine research, China holds notable potential for growth. Large-scale Chinese cohort studies have contributed to the understanding of fetal growth restriction and hypertensive disorders. Given the vast population and increasing emphasis on prenatal screening in China, there is strong potential to enhance its global research contributions. Strengthening international collaborations and focusing on large-scale studies could further propel the impact of China in the field.3 More than 800 institutions contributed to the journal during the study period. The five most productive institutions were King’s College Hospital London (208 publications), St George’s University London (112), University of Exeter (96), Katholieke Universiteit Leuven (79), and University of Barcelona (77) (Supplementary Fig. 2B, https://links.lww.com/MFM/A133). At the author level, Nicolaides KH (147 publications), Khalil A (50), Poon LC (43), Youssef A (20), and Cortes Sanz M (18) were the most prolific contributors. Taken together, these findings demonstrate clear academic clustering, with a small number of countries, institutions, and investigators contributing disproportionately to the journal’s output. This pattern likely reflects long-established fetal medicine networks, editorial visibility, access to high-volume clinical datasets, and sustained leadership in multicenter screening studies3–6 (Supplementary Fig. 2B, https://links.lww.com/MFM/A133). Collaboration patterns Collaboration network analysis further highlighted the central role of the UK. At the country level, the UK showed the strongest international linkages, especially with the United States, the Netherlands, Italy, Belgium, and Spain. The strongest bilateral collaboration was between the UK and Belgium. At the institutional level, the most prominent collaborative ties were observed among King’s College Hospital London, University of Exeter, Medway Maritime Hospital, and King’s College London. At the author level, Nicolaides KH showed strong and stable collaboration with Syngelaki A, Akolekar R, Wright A, and Wright D. These findings suggest that UOG research output is not merely concentrated geographically but is also organized around highly interconnected research hubs. Such hubs appear to combine clinical cohorts, epidemiologic modeling, and translational screening frameworks, thereby reinforcing their academic influence6,7 (Supplementary Fig. 3, https://links.lww.com/MFM/A133). Keyword structure and thematic evolution After keyword cleaning and standardization, 3333 keywords remained for co-occurrence analysis. The results showed that “pre-eclampsia” was the most prominent keyword and had strong links with “placental growth factor,” “mean arterial pressure,” “uterine artery Doppler,” and “early pregnancy screening.” This pattern reflects the sustained emphasis on first-trimester prediction models combining maternal characteristics, ultrasound parameters, and angiogenic biomarkers.4,6,8 Another notable cluster linked “fetal growth restriction” with “small-for-gestational-age,” reflecting a continued focus on differentiating physiologically small fetuses from those with pathologic growth restriction through more refined ultrasound phenotyping and risk modeling.7,9 These two thematic domains—placenta-related hypertensive disease and fetal growth disorders—appear to form the core clinical axis of recent UOG literature(Supplementary Fig. 3, https://links.lww.com/MFM/A133). Keyword clustering identified 11 major themes, including small-for-gestational-age, mean arterial pressure, fetal surgery, fetal echocardiography, transperineal ultrasound, trisomy 21, artificial intelligence, cervical length, monochorionic twins, cesarean delivery, and second trimester. The clustering quality was robust (modularity Q = 0.6302; silhouette S = 0.8623), indicating that the thematic structure was stable. Time-zone visualization showed a gradual shift in emphasis over the decade. Earlier years were characterized by conventional prenatal screening themes, such as trisomy 21 risk assessment and diagnostic accuracy studies. Over time, these were increasingly complemented by more advanced topics, including competing-risk models, fetal therapy, placental disorders, and AI-assisted image analysis. In the AI-related cluster, there was an evolution from terms such as “diagnostic accuracy” to “deep learning,” “machine learning,” and “neural networks,” suggesting that ultrasound research is moving beyond image interpretation alone toward computational risk stratification and automated decision support8,10 (Supplementary Fig. 4 and Supplementary Table 1, https://links.lww.com/MFM/A133). Emerging hotspots Burst analysis highlighted several keywords with strong temporal signals. The most intense burst term was “Pyramid of pregnancy care,” consistent with the growing adoption of early, risk-stratified prenatal care models. The longest-lasting burst terms included “pelvic floor,” “competing-risks model,” “machine learning,” and “fetal brain.” More recent burst terms included “angiogenic factor,” “placental insufficiency,” “artificial intelligence,” and “fetal brain.” These findings suggest that contemporary UOG research is increasingly organized around two intersecting trajectories—the refinement of early pregnancy prediction frameworks for pre-eclampsia, fetal growth restriction, and related placental disorders, and the expansion of intelligent imaging approaches that integrate ultrasound with computational analytics. The convergence of these trajectories may support future multimodal assessment systems that combine imaging, biomarkers, and algorithmic modeling8,10 (Supplementary Fig. 5, https://links.lww.com/MFM/A133). Discussion This journal-focused bibliometric analysis provides a concise picture of how UOG literature evolved between 2015 and 2025. Three broad observations emerge. First, the field has shifted from largely descriptive and morphology-based ultrasound applications toward integrated prediction paradigms. This is especially evident in pre-eclampsia and fetal growth restriction, where traditional sonographic markers are increasingly combined with maternal factors, angiogenic biomarkers, and competing-risk or machine-learning models.4,6,8 Second, the academic landscape remains highly centralized. The UK occupies a dominant position not only in publication output but also in collaborative connectivity. Key institutions such as King’s College Hospital London, St George’s University London, and the University of Exeter serve as major hubs, reflecting the central role of well-established fetal medicine networks in advancing the field. The most highly connected authors are predominantly from fetal medicine or maternal-fetal medicine, underscoring the field’s significant influence on the journal’s output. Third, AI has transitioned from a peripheral methodological topic to a central research frontier, particularly in predictive models for pre-eclampsia and fetal growth restriction. Breakthroughs around 2020–2021 were driven by advancements in imaging and computational analysis. Keywords such as “machine learning” and “neural networks” now dominate the field, reflecting the growing role of AI in prenatal screening. Institutions such as King’s College Hospital London have integrated AI-driven ultrasound analysis with traditional methods, thereby improving diagnostic accuracy and enabling personalized care. This keyword evolution indicates that there has been a shift from proof-of-concept diagnostic applications to broader clinical integration, such as automated image interpretation, gestational assessment, and risk modeling.8,10 While terms related to AI remain less frequent than keywords related to placenta and screening, their sustained burst pattern indicates that this technology is exerting an ongoing influence in the field. This study also illustrates the value of a journal-targeted bibliometric strategy. Because obstetric and gynecologic ultrasound research is marked by substantial terminology fragmentation, broad topic-based retrieval may omit important studies or introduce notable thematic heterogeneity. Focusing on a leading specialty journal provides a more homogeneous and clinically coherent dataset, allowing for a clearer characterization of high-visibility trends within the field. Several limitations should be acknowledged. First, the analysis was restricted to articles published in UOG and indexed in the Web of Science Core Collection; accordingly, the findings are not generalizable to all obstetric and gynecologic ultrasound literature. Second, bibliometric results are influenced by indexing practices, database updates, and citation time lag, particularly for recently published work. Third, despite keyword cleaning and standardization, some subjective judgment in term merging was unavoidable. In conclusion, UOG literature over the past decade reflects a clear transition toward precision screening, placental and fetal risk stratification, and intelligent ultrasound applications. Pre-eclampsia, fetal growth restriction, angiogenic biomarkers, and AI-driven analysis represent the major interconnected hotspots. As maternal-fetal medicine moves toward more integrated and data-intensive care pathways, continued attention to these themes may help guide both future research and clinical translation.