The database known as the Web of Science Core Collection (WoSCC) is widely used in bibliometric research and encompasses the Science Citation Index Expanded (SCIE), Social Science Citation Index (SSCI), and Emerging Sources Citation Index (ESCI). Data was obtained from the WoSCC database on September 23, 2024, to avoid any potential deviation due to daily updates. The search strategy used was [TS = (medical education AND neurology)] AND [Language = (English)], limiting the publication years to between 2000 and 2023. The results obtained from the search were saved in “Full Record and Cited References” and “Plain Text” formats. For further analysis, the downloaded files were renamed to “download_^*.txt,” making them suitable for use with CiteSpace software. Notably, 918 documents comprising 10 document types were obtained. To enhance the significance of the data, the documents that are irrelevant to our restrictions were removed: (1) 12 of publication year which are unqualified, (2) 4 of document types are correction (n = 1), book chapter (n = 1) and Biographical-Item (n = 2). Therefore, 900 documents were finally collected to further research (Figure 1). The document types of each paper were listed in Table 1, the number of Articles have the highest count and H-index (n = 724, H-index = 49), followed by Review Articles (n = 82, H-index = 21) and Editorial Material (n = 66, H-index = 11).

We employed CiteSpace 6.2.R4 (developed by Chaomei Chen), VOSviewer 1.6.20 (created by Nees Jan van Eck and Ludo Waltman), alongside Microsoft Excel 2019, to conduct our bibliometric analysis and visualization. The initial phase involved data cleaning wherein terms such as “alzheimer disease” and “alzheimers-disease” were consolidated into “alzheimers-disease,” and “eeg (electroencephalography)” was standardized to “eeg (Paunkov et al., 2019).”

CiteSpace serves as an effective tool for bibliometric and visual analysis, particularly in identifying collaboration patterns, significant themes, structural elements, prospective trends, and dynamics within scientific disciplines. Therefore, we leveraged CiteSpace to scrutinize the co-occurrence of different nations, regions, and institutions. Additionally, we looked into the dual-map of journals, timelines of references, citation bursts, keyword evolution, and keyword bursts. The following parameters were employed: timespan (2000-2023), slices per year (1), no pruning, selection criteria (Top N = 100), and a minimum burst duration of 2 years. Cluster labels were derived through light semantic indexing (LSI) and the log-likelihood ratio (LLR) method, while other settings remained as default. Within the CiteSpace visualization, node sizes correspond to the frequency of co-occurrences, and the connecting lines highlight the co-occurrence relationships. Node and line colors vary by year, shifting from gray to red to represent the timeline from 2000 to 2023. Nodes represented by purple circles indicate a high betweenness centrality (≥0.10), acting as connectors among different networks (Song C. et al., 2024).

VOSviewer is a bibliometric tool that excels in creating and visualizing knowledge maps, illustrating different types of clusters, overlays, or density colors. It has been utilized to analyze the co-occurrence of authors alongside co-cited authors, journals with their co-cited counterparts, as well as co-cited references and keywords. We employed full counting as the method for data representation; other thresholds are detailed in the corresponding chapter. In the cluster map, the node size corresponds to co-occurrence frequencies, while identical colors denote the same cluster. Additionally, links illustrate co-occurrence relationships, with the thickness reflecting a computed strength value, which is directly related to the number of joint publications by two researchers or the frequency of two keywords appearing together. In density maps, the size of words and the brightness of yellow are positively correlated with co-citation frequency. The overlay map uses color to signify the average publication year.

To analyze annual publications, we utilized Excel software. Additionally, we retrieved the impact factor (IF), journal citation reports (JCR) for various journals, and the H-index for scholars from the Web of Science on September 23, 2024.

We screened 900 articles from the WoSCC database for analysis (Figure 1, Supplementary Table 1). The overall number of articles from 2000 to 2010 was still relatively small, but the number of articles increased gradually during the period from 2011 to 2023 (Figure 2).

Among the total of 900 articles, 724 are research articles and 82 are reviews, which together account for the majority of the publications (Table 1). These articles originate from 92 countries/regions and 893 institutions (Supplementary Tables 2, 3). The United States has the highest number of publications (N = 513), followed by England (N = 77) and Canada (N = 70). However, Canada's centrality is less than 0.1, suggesting that it may not serve as a central node in research related to neurology medical education (Figure 3A). The top 10 contributors are listed in Table 2. In contrast, the United States exhibits a higher centrality (centrality = 0.36). The density of country/region co-occurrence is 0.138, reflecting a strong connection among them. Harvard University is identified as the most productive institution, although its centrality is relatively low (n = 74, centrality = 0.03), followed by Mayo Clinic (n = 57, centrality = 0.06) and the University System of Ohio (n = 56, centrality = 0.03). In contrast, King's College Hospital (n = 5, centrality = 0.09) and the University of London (n = 23, centrality = 0.08) demonstrate higher centrality (Figure 3B, Table 2).

A total of 4,339 authors participated in the research on neurology medical education (Figure 4A; Supplementary Table 4). The top ten authors collectively published more than six articles, while 74 authors produced at least three papers; however, 28 of these authors are not interconnected (Figure 4A). Gilbert Donald L from Cincinnati Children's Hospital Medical Center authored the highest number of neurology-related articles (n = 10), followed by Lukas Rimas V and Soni Madhu (Table 3). Figure 4A features seven colors, each representing a distinct cluster among the authors. Active collaborations typically occur within the same cluster, as seen with Sarva Harini and Weber Daniel. Additionally, there are collaborations between different clusters, exemplified by Albert Dara V. F. and Gilbert Donald L.

Co-cited authors are authors who are cited in an article. Among the 17,656 cited authors, five have been cited more than 50 times (Figure 4B, Supplementary Table 5). Figure 4B is presented as a density plot, effectively illustrating high-frequency co-cited authors, with warmer colors indicating a greater number of citations. As detailed in Table 3, Figure 4B, Jozefowicz Rf, Schon F, and Zinchuk Av were the most frequently co-cited authors.

Due to the inherent limitations of CiteSpace and VOSviewer visualizations, these figures cannot encompass all available information. Consequently, we provide the complete data in Supplementary Table 5, as well as in the figure below.

A total of 297 journals published research on the relationship between neurology and medical education, with the top 15 journals contributing 390 articles, which accounts for 43.33% of the total (see Table 4, Supplementary Table 6). The journal with the highest number of publications is Neurology (n = 118), followed by BMC Medical Education (n = 35) and Pediatric Neurology (n = 26). Among 7,130 co-cited journals that received a total of 26,241 citations (Supplementary Table 7), 28 journals had more than 100 citations. Neurology (n = 1,310), Academic Medicine (n = 829), and JAMA-Journal of the American Medical Association (n = 437) received the highest number of citations. Furthermore, the top 15 co-cited journals accounted for 22.98% of all cited sources (see Table 4).

The dual-map overlay demonstrating the distribution of topics within academic journals is depicted in Figure 5. On the left side, the citing journals are displayed, while the cited journals are featured on the right, with colored lines representing the relationships of citations. As shown in Figure 5, there are two main citation paths: one originating from journals in Neurology, Sports, and Ophthalmology leading to those in Health, Nursing, and Medicine, and another stemming from Psychology, Education, and Social journals. The first path shows a notable frequency of citations, with a high f-value of 62,890, whereas the second path exhibits a strong citation relationship indicated by its z-value of 4.1560464. Moreover, there are two significant citation pathways from journals in Medicine, Medical, and Clinical fields to both Health, Nursing, and Medicine journals and to Psychology, Education, and Social journals. Figure 5 illustrates that both Health, Nursing, and Medicine journals along with Psychology, Education, and Social journals receive citations from both Neurology, Sports, and Ophthalmology journals, as well as from Medicine, Medical, and Clinical journals.

Out of the 22,574 references cited, 38 were mentioned a minimum of 10 times (refer to Supplementary Table 8). As shown in Table 5, the ten most co-cited works received at least 20 mentions. Remarkably, the most frequently co-cited work is a letter authored by Jozefowicz RF et al., published in the Archives of Neurology in 1994 (n = 83) (Jozefowicz, 1994). Additionally, among the top 10 references, seven were articles, while the remaining included letters, editorial pieces, and reviews. The reference timeline view illustrates the evolution of research hotspots over time. The most frequent items in each cluster are designated as cluster labels, and the remaining items are detailed in Supplementary Table 9. As shown in Figure 6, cluster #7 (psychiatry education/gross anatomy) emerged earlier, whereas clusters #1 (medical education/neuroanatomy education), 2 (mental health/podcasting), 4 (cognitive heuristics/neurosurgery), 6 (clinical neurophysiology/online teaching), and #8 (.../virtual learning) are ongoing, which can be regarded as novel-edge research.

References exhibiting citation bursts are those that have been cited significantly more frequently over a specific period. A total of 265 references were identified as having citation bursts, with the top 20 listed in Figure 7. The strongest burst (strength = 5.76) occurred in a paper titled “Structure of neuroscience clerkships in medical schools and matching in neuromedicine,” published in Neurology by Dara V Albert in 2015 (Albert et al., 2015), which experienced a citation burst from 2016 to 2018. Additionally, there are no references currently experiencing a burst.

A total of 2,993 keywords were extracted, with 106 keywords appearing at least 10 times and 40 keywords appearing at least 20 times (see Supplementary Table 10). Table 6 indicates that 'education' was the most frequent keyword (n = 211), followed by “neurology” (n = 191) and “medical education” (n = 133). The keywords most commonly associated with neurological diseases and medical education include “neurophobia” (n = 68), “epilepsy” (n = 47), and “stroke” (n = 42).

Figure 8A presents a high-frequency keyword overlay map (n ≥ 15), where the color represents the average year of publication. Notably, 'neuroanatomy,' 'anatomy,' 'neurosurgery,' 'intervention,' and 'people' are emerging fields, indicated in yellow. The timeline view in Figure 8B illustrates the top two high-frequency keywords in each cluster over time. It is evident that nine of the ten clusters (excluding #7) are still active. Among these, cluster #0 (medical education/neuroanatomy education) is the largest, followed by cluster #1 (knowledge/critical care), cluster #2 (neurology residents/students), and cluster #3 (graduate medical education/resident education). Additional information is provided in Supplementary Table 11.

Keyword bursts refer to those that have been cited with significant frequency over a specific period. As illustrated in Figure 9, “receptor neurophobia” exhibited the strongest burst (strength = 5.28), followed by “medicine” (n = 3.65) and “American academic” (n = 3.47). Notably, “resident education,” “medical education and training,” “developmental neurology & neurodisability,” and “parents” continued to show bursts of activity through 2023.

Based on data from the WoSCC database spanning from 2000 to 2023, a total of 900 articles exploring the relationship between neurology and medical education were published across 297 academic journals. These articles were authored by 4,339 individuals from 893 institutions representing 92 countries and regions. The growing body of research indicates an increasing interest in the association between neurology and medical education. The investigation of neurology and medical education began in 1950 when Woltman reported on the neurological aspects of the Minnesota experiment in graduate medical education (Woltman, 1950). Since then, research in this area has progressed steadily. Over the past decade, the number of publications on neurology and medical education has significantly increased, with the volume of articles published between 2010 and 2023 nearly quadrupling compared to those published between 2000 and 2010.

In analyzing country and regional involvement, the number of published articles and centrality are key factors to consider. High centrality nodes (≥0.10) indicate the 'bridge' effect these countries and regions have on the global cooperation network. Table 2, Figure 3 illustrate that the United States, England, and Canada lead in research on neurology and medical education. Among the top 10 institutions publishing research projects, all are located in the United States. Notably, Canada has centrality values below 0.1, while the United States and England have centrality values of 0.36 and 0.12, respectively, suggesting their dominant positions in neurology and medical education research. Furthermore, the United States, England, Australia, and the Netherlands all have centralities exceeding 0.1, underscoring their significant roles in global research on neurology and medical education. Overall, active collaboration among these countries and institutions is evident in terms of network density. Therefore, strengthening international cooperation with the United States in medical education is conducive to the development of neurology teaching.

Among the top 10 authors and co-cited authors, Donald L Gilbert has published the most articles on neurology and medical education research, underscoring his significant contributions to these fields. Gilbert, a Full Professor at Cincinnati Children's Hospital Medical Center, specializes in pediatric neurological diseases, including movement disorders and Tourette Syndrome. He has also identified common mechanisms linking transcranial magnetic stimulation and behavioral disorders. Notably, one of his studies published in the journal Pediatric Neurology in 2016 demonstrated that the requirement of 12 months of adult neurology training for Child Neurology and Neurodevelopmental Disability certification does not align with the views of the majority of program directors (Valencia et al., 2016). Jozefowicz Ralph F. is the leading co-cited author, specializing in medical education and neuromuscular disease at the University of Rochester. His research primarily focuses on neurology education, residency training in neurology, and the education of medical students. In 1994, Jozefowicz published a study in Archives of Neurology that investigated the concept of neurophobia, which he defined as a fear of the neural sciences and clinical neurology stemming from students' inability to apply their knowledge of basic sciences to clinical situations. These negative sentiments are relevant to the study of neurology in medical school (Jozefowicz, 1994).

The journal Neurology is recognized for publishing a substantial amount of research in the fields of neurology and medical education, and it is distinguished by having the highest number of co-citations. This assertion is supported by the fact that three of the top ten highly cited papers were published in Neurology and BMC Medical Education. These journals primarily concentrate on cellular neurology, medical education, and related disciplines. Furthermore, Neurology ranks first in the number of articles published, highlighting its significance in the dissemination of research concerning neurology and medical education. Dual-map analysis indicates that research in these areas is predominantly centered around psychology, nursing, education, and social sciences.

The set of references that are frequently cited together within a specific research community may reflect the collective knowledge base. Among the top ten cited references, six primarily address neurophobia in neurological education, emphasizing the importance of confronting this phenomenon, along with its implications and potential solutions. Additionally, two articles focus on the neurology clerkship: one presents a core curriculum outlining the clinical neurology skills and knowledge necessary for adequate preparation (Gelb et al., 2002), while the other evaluates the status of the neurology clerkship in U.S. medical schools and its impact on graduates entering the field (Albert et al., 2015). These findings suggest that neurophobia significantly influences medical education in neurology through its effects on medical students.

Regarding burst analysis, the absence of references currently in burst indicates a need for further study on the intersection of neurology and medical education. The strongest burst suggests a neurology clerkship and highlights opportunities for students to connect with peers who are matching neuromedicine residencies. Two references from 2020 are noteworthy: one discusses evidence-based recommendations and educational interventions aimed at addressing neurophobia in medical education, while the other establishes internationally recognized standards and provides guidelines related to neuroanatomical knowledge.

Nowadays, it is very important for researchers to quickly understand current research hot spots. Therefore, bibliometrics provides a quantitative method that can reflect research hot spots in the academic field through keyword co-occurrence. Overlap and timeline views can reflect the development of new research, and reference clusters and citation bursts can represent emerging topics. This study attempts to objectively evaluate through keyword co-occurrence (Tables 5, 6), keyword overlap and timeline (Figure 8), keyword burst (Figure 9), reference timeline (Figure 6), reference burst (Figure 7), etc. Neurology and hot spots and trending medical education research.

This study represents the first bibliometric analysis systematically examining publications in neurology and medical education over the past two decades. Unlike traditional reviews, bibliometric analysis offers a novel and objective perspective on the evolving trends within this research field. We employed various bibliometric software tools for analysis, yielding more results across multiple dimensions (Shen et al., 2022). This study aims to enhance public understanding of the significance of neurology in medical education, provide scholars with a comprehensive overview of research in this area, and offer objective guidance for the future development of neurology and medical education research. However, this study has certain limitations. Firstly, we exclusively searched the WoSCC database for published English articles, thereby excluding non-WoSCC and non-English publications. Nonetheless, English articles in WoSCC are the most frequently utilized data source in bibliometrics and can represent a substantial portion of the relevant information. Secondly, the bibliometric methodology relies on natural language processing, which may introduce biases, as noted in other bibliometric studies (Yan et al., 2021). Despite these limitations, our findings provide a more comprehensive and objective overview of the field. Furthermore, the sample reflects the current volume of research on neurology and medical education, indicating that this area is an emerging direction for future inquiry.