The wide coverage of the WoS Core Collection (WoSCC) database, which provides rich citation analysis tools, indicators, and powerful visualization capabilities, is the most commonly used database for bibliometric analysis (Birkle et al., 2020). Previous studies have convincingly demonstrated the efficacy of bibliometric analysis conducted on the WoSCC database (Wang and Maniruzzaman, 2022; Li et al., 2023). By employing this tool, we set out to retrieve relevant literature published between January 1, 2012, and December 31, 2022. The main topics of data retrieval were “neuromodulation techniques” and “disorders of consciousness.” Therefore, the search terms were as follows: TS = (“neuromodulation techniques” OR “deep brain stimulation” OR DBS OR “Spinal cord stimulation” OR SCS OR “transcranial direct current stimulation” OR tDCS OR “transcranial magnetic stimulation” OR TMS OR “vagus nerve stimulation” OR VNS OR “median nerve stimulation” OR MNS) AND TS = (“disorders of consciousness” OR “disturbance of consciousness” OR “consciousness disorder” OR consciousness).

After screening the titles and abstracts, we selected studies that utilized neuromodulation techniques for treating disorders of consciousness. The document type was only limited to articles and reviews. Other irrelevant literature was excluded, including meeting abstracts, letters, editorial material, book chapters, non-English papers, etc. Additionally, the duplicated articles have been eliminated. Before analysis, two researchers independently screened the data. Any discrepancies were discussed and resolved by a third reviewer. The flowchart of the inclusion criteria is presented in Figure 1. Finally, a total of 338 records were retrieved for the purpose of conducting a bibliometric analysis.

Data screened from the Web of Science were analyzed using CiteSpace (version: 6.2.R4; Drexel University, USA) and Microsoft Excel 2019. The parameters for CiteSpace were set with a “Time Sliding” value of 1 year, and the node type was selected based on the analysis.

In this study, CiteSpace was used to analyze the dual-map overlap of journals, cluster view, burst detection of cited literature, and centrality of country, institution, and author. The dual-map overlay visualization presents a graphical representation of how articles are distributed, how citation patterns are evolving over time, and how the center of gravity is shifting in different disciplines. The distributions consist of citing journals positioned on the left and cited journals positioned on the right. The citation line represents the context of the citation. The Z-score and F-score provide more consistent and robust trajectory maps, enabling the identification of significant citation paths in the dual-map. Cluster view is conducted on the generated map, where each cluster is identified through citations that include the title, keywords, and subject headings from the abstract of the citing reference. The purpose of burst detection is to identify significant changes in citation numbers over a given period, making it useful for tracking keyword trends. The index of betweenness centrality is used to determine the significance of nodes in a network.

Microsoft Excel 2019 was employed to display publication features and analyze annual publication and citation trends. The data were extracted from Science Citation Index-Expanded of the WoSCC Database, formatted as annual publications, citations, subject categories, countries, institutions, authors, and journals. Besides, the 2022 impact factor (IF), Hirsch index (H-index) and Journal citation reports (JCR) division of journals were also included. IF gages the rank or significance of a journal by calculating the mean number of citations accorded to its selected papers in a given year. The H-index is a metric that measures the scientific impact of an author/country by considering the minimum number of papers (H) that must be published, with each paper receiving at least H citations. JCR is an indexing and evaluation tool that assesses the quality and impact of journals based on their citation analysis data.

From 2012 to 2022, a total of 338 articles were analyzed, including 277 articles (81.95%) and 61 reviews (18.05%). Figure 2 illustrates the distribution of annual publications and citations related to neuromodulation techniques in DOC research. We have identified two trend lines, represented by orange and blue. The orange line signifies the overall trend in the broader field, while the blue line indicates the trend in Neurosciences. In terms of annual publications (Figure 2A), the overall trend in the field has undergone two distinct phases over the past decade. Initially, there was a gradual increase in publication numbers between 2012 and 2015, followed by an acceleration leading to a peak between 2016 and 2022, exhibiting a consistent upward trajectory year on year. However, the growth rate in Neurosciences was slower compared to the overall trend, attributed to the less reliable exponential growth model (R2 = 0.7656). Figure 2B illustrates the annual citation distribution of the studies included. The overall trend of citation counts has steadily risen, increasing from 15 in 2012 to 1781 in 2022. Additionally, the growth in Neurosciences aligned with the overall growth. By applying an exponential growth model to evaluate the correlation between citation counts and publication years, the results suggest that the model is consistent with the trend of annual citation volumes (Orange: R2 = 0.9797; Blue: R2 = 0.9726).

According to Figure 3, it was observed that the highest average number of citation per paper (n = 106.38), and citations (n = 2,234) were recorded in 2014. The highest H-index (n = 19) was observed in 2017. Furthermore, the largest number of published articles (n = 59) and open access (n = 47) occurred in 2022.

The 338 studies that were included in this analysis were published across 176 different academic journals. Table 1 displays the details of the top 10 journals. Frontiers in Neurology published the most number of papers (n = 16), and the highest open access (n = 16), followed by Brain Sciences (n = 13), and Frontiers in Neuroscience (n = 13). In terms of cited frequency, Clinical Neurophysiology ranked first (n = 2,303), followed by PLOS ONE (n = 396) and Brain Stimulation (n = 361). Brain Stimulation also presented with the highest Impact Factor (IF 2022 = 7.7). Frontiers in Neurology and Brain Stimulation had the highest H-index (n = 10).

The dual-map overlaps of the journals are presented in Figure 4, indicating the reciprocal relationship between cited and citing journals. We identified four main citation trajectories (pink and orange), where journals in neurology, sports, and ophthalmology (pink trajectory) were significantly more frequently cited by Molecular, Biology, Genetics (Z = 5.82, f = 1,515), and Psychology, Education, Social (Z = 5.45, f = 1,468) fields. In addition, journals in Molecular, Biology, Immunology (orange trajectory) were influenced by journals in Molecular, Biology, Genetics (Z = 2.74, f = 756), and Psychology, Education, Social (Z = 2.34, f = 655) fields.

The 338 publications were sorted into 64 different WoS subject categories. We conducted an analysis of the top 10 published disciplines (Figure 5). Neurosciences led with the highest number of publications (n = 170), open-access value (n = 1,221), citations (n = 5,772), and H-index (n = 30), followed by Clinical Neurology (n = 100) and Multidisciplinary Sciences (n = 27). Medicine Research Experimental had the highestcitation per paper (n = 61.33).

Based on the analysis of reference co-citation, the research categories were segmented into 11 groups (#0–11). The timeline view of clusters for citation information of the cluster domains is shown in Figure 6. The largest cluster (#0) had 46 members, which was labeled as conscious state by Latent Semantic Indexing (LSI). The major citing article of the cluster was “Assessing consciousness in coma and related states using transcranial magnetic stimulation combined with electroencephalography.” The second-largest cluster (#1) had 43 members labeled as prolonged disorder by LSI. The major citing article of the cluster was “Electromagnetic brain stimulation in patients with disorders of consciousness.”

The Sigma value is calculated using a formula that combines both centrality and burstness metrics to identify innovative topics. The formula for Sigma is given as follows: Sigma = (centrality+1) burstness. We have summarized the top 3 innovative references (Table 2). Two articles were observational studies and one article was a randomized controlled trial.

Figure 7A shows the top 10 countries based on the number of publications of neuromodulation techniques in DOC research. The USA had the highest number of publications (n = 105), open access papers (n = 81), and H-index (n = 30). Italy had the greatest number of citations (n = 5,224). Germany ranked the highest number of citation per paper (n = 128.39). Figure 7B shows the top 10 countries with the strongest citation burst. Italy represented the strongest citation burst (strength = 2.2) from 2012 to 2013, followed by Russia (strength = 1.17) and Turkey (strength = 1.08).

Figure 8A shows the top 10 institutions based on the number of publications of neuromodulation techniques in DOC research. The University of Liege has the highest number of publications (n = 41), open access value (n = 36), and H-index (25). The University of Milan ranked the highest number of citations (n = 2,815) and citation per paper (n = 122.39). Figure 8B shows the top 10 institutions with the strongest citation burst. Yanshan University represented the strongest citation burst (strength = 3.36), maintaining a high-intensity outbreak in 2017–2018.

Figure 9A shows the top 10 authors according to the number of publications of neuromodulation techniques in DOC research. Laureys Steven had the largest number of papers (n = 38), open access papers (n = 32), citations (n = 2,681), and H-index (n = 23). Massimini Marcello had the greatest number of citation per paper (n = 113.4). Figure 9B shows the top 10 authors with the strongest citation burst. Calabro, and Naro had the strongest citation burst (strength = 4.15), with the burst lasting 2 years (2015–2016).

The collaboration maps for various countries, institutions, and authors are shown in Figure 10. Regarding centrality, the top three countries were Italy (n = 0.32), the USA (n = 0.22), and England (n = 0.13). The top three ranked institutions by centrality were Sichuan University (n = 0.24), Capital Medical University (n = 0.23), and Institut National de la Sante et de la Recherche Medicale (n = 0.21). The top three centrality authors were Laureys Steven (n = 0.03), Ruffini Giulio (n = 0.02), and Rossi Simone (n = 0.02). Interestingly, we observed that Laureys Steven and Rossi Simone primarily conducted research on the application of transcranial magnetic stimulation together with high-density electroencephalography to assess brain connectivity in disorders of consciousness (Gosseries et al., 2014; Ragazzoni et al., 2017), while Ruffini Giuli’s investigation concentrated on examining the impact of tDCS on patients with consciousness disorders (Martens et al., 2020).

Figure 11 presents the top 25 keywords with the strongest citation bursts from 2012 to 2022. The keyword with the highest burst value was coma recovery scale (n = 3.99), while the keywords with the longest burst period were effective connectivity, human motor cortex, response, and somatosensory evoked potentials, lasting 5 years. At the conclusion of 2022, the most prevalent keywords among cited publications from 2020 to 2022 encompassed unresponsive wakefulness and rtms.

The top 10 most cited papers are listed in Table 3. The citations of these articles ranged from 147 to 1,252. Three of them have been cited more than 500 times. The article entitled “Evidence-based guidelines on the therapeutic use of repetitive transcranial magnetic stimulation (rTMS)” published by Lefaucheur et al. in 2014 in Clinical Neurophysiology has been cited the most (1,252 citations) (Lefaucheur et al., 2014). Four of the top 10 articles were published in journals with an impact factor of 10 or higher, including Science Translational Medicine, Brain, Neurology, and Lancet Neurology.

Based on our research results, the annual publication volume showed a continuous but irregular annual growth trend, with the most significant growth trend occurring between 2016 and 2022. However, the growth rate in Neurosciences was slower than the overall trend. Furthermore, the total number of citations for articles also showed a continuous annual growth trend, increasing from 15 to 1,781. The growth of Neurosciences was consistent with the overall growth. These results indicate that the global attention to neuromodulation techniques in DOC has been steadily increasing.

In terms of authoritative journals, Frontiers in Neurology (n = 16), Brain Sciences (n = 13), and Frontiers in Neuroscience (n = 13) were ranked as the top three. Among the top 10 journals, only two journals belong to Q1. Furthermore, only Brain Stimulation had an IF exceeding 5, providing a platform for high-impact neuromodulation research from an international perspective. It is thus clear that the quality of studies investigating the role of neuromodulation techniques for DOC still needs to be strengthened. In addition, by examining the dual-map overlaps of published journals to gain insights into the citation trajectories between different fields, it could be found that the neuromodulation techniques in DOC research were influenced by a wide range of fields from molecular, biology, and genetics to psychology, education, and social. These results indicate that the neuromodulation techniques in DOC research have a cross-disciplinary nature, and future research should pay more attention to the journal dynamics in these fields.

Among authoritative countries, the USA (n = 105) made the most contributions to this research domain, followed by China (n = 80) and Italy (n = 75). However, due to the relatively low citation frequency of most Chinese literature, the average number of citations per article was low. This phenomenon reflects the lack of overall research quality of Chinese publications and the fact that they have not yet achieved widespread dissemination in the field. As for authoritative institutions, the University of Liege ranked first (n = 41) in terms of the number of publications, and it is a leading European research university alliance. Additionally, this institution also exhibited the highest open-access value (n = 36), and H-index (n = 25). In the realm of authoritative authors, Laureys Steven and Gosseries Olivia stood out in this field with exceptional accomplishments, publishing over 20 papers and accumulative 2000+ citations. Moreover, Laureys Steven also exhibited the highest centrality (n = 0.03). From the perspective of the cooperative network, Italy had the greatest centrality (n = 0.32), followed by the USA (n = 0.22) and England (n = 0.13). Notably, the USA and Italy exhibited significant centrality and publication output, suggestive of occasional dual affiliations among contributors from these countries. Sichuan University (n = 0.24) was the top institution with the highest centrality, followed by Capital Medical University (n = 0.23) and Institut National de la Sante et de la Recherche Medicale (n = 0.21). Concurrently, the majority of the top 10 institutions with the highest centrality and publication output were from China and the USA. However, these institutions primarily collaborated with other institutions within their respective countries. It is not difficult to find that the cooperation between countries and organizations remains regional. Therefore, it is imperative to enhance collaboration and communication with domestic and foreign research institutions in the future.

According to the subject categories of neuromodulation techniques in DOC research, Neurosciences (n = 170) ranked first in this field, followed by Clinical Neurology (n = 100) and Multidisciplinary Sciences (n = 27). The top 10 discipline categories were Neurosciences, Clinical Neurology, Multidisciplinary Sciences, Rehabilitation, Surgery, Psychiatry, Medicine General Internal, Behavioral Sciences, Medicine Research Experimental, Psychology, suggesting neuromodulation techniques for DOC are a complex medical challenge that necessitates multidisciplinary communication and cooperation.

In terms of reference analysis, the earliest of the 10 most cited articles was published in 2012, while the latest was published in 2019, where, BRAIN (Rosanova et al., 2012) and, LANCET NEUROL (Thibaut et al., 2019) were defined as classic literature by the bibliometric method, highlighted their significant academic value in this field. Additionally, the most relevant citer to the largest cluster was “Assessing consciousness in coma and related states using transcranial magnetic stimulation combined with electroencephalography.” This review pointed out that TMS-EEG has great potential in identifying consciousness markers at the individual level, and may be of great value for clinicians in assessing consciousness (Gosseries et al., 2014). Moreover, through examination of the three innovative studies, by Thibaut et al. (2014), Rosanova et al. (2012) and Ferrarelli et al. (2010), it was found that NINT, especially TMS and tDCS, played a significant role in the evaluation and management of DOC. NINT seems to have more advantages in treating disorders of consciousness, but the reasons are complex. INT often requires surgery or implantable devices, which face the risks of infection, surgery, and device failure (Zaghi et al., 2009). NINT relies on external stimuli to regulate the nerves, which is subject to the accuracy of the device, operating skills, and the selection of stimulation parameters. In human studies, ethical considerations are also important, as INT must comply with medical ethics and fully consider potential conflicts of interest (Giacino et al., 2012), such as the relationship between researchers and device manufacturers. Future research should explore various electrophysiological assessments and treatments that offer new possibilities for enhancing the quality of life of patients with DOC.

In the domain of bibliometrics, keywords hold a significant position in summarizing articles. Through keyword analysis, we can deeply explore and reveal emerging research trends, and provide targeted guidance for future research. “coma recovery scale” was the keyword with the highest burst value, and was also the core content of assessment in DOC. Furthermore, in terms of count numbers and centrality, the top three keywords were UWS/VS, TMS and MCS. In addition, the keywords with the most outbreaks of cited publications included unresponsive wakefulness and rTMS by the end of 2022. Based on the above results, “UWS/VS,” “MCS” and “TMS” indicate potential research hotspots and frontiers. We now proceed to a detailed examination of these potential hotspot terms.