Bibliometric analysis of original studies on stereoelectroencephalography from 1990 to 2023

Abstract

BACKGROUND

Stereoelectroencephalography (SEEG) is a minimally invasive preoperative evaluation of drug-resistant focal epilepsies. Increasing preclinical data show that SEEG has tremendous diagnostic and therapeutic value for epilepsy. There are currently many studies on SEEG.

AIM

To summarize the development and application of SEEG.

METHODS

We conducted a bibliometric analysis of research on SEEG from 2019 to 2023. We obtained publications on SEEG from the Web of Science Core Collection (WoSCC) database. Excel, VOSviewer, and CiteSpace were responsible for the analyses. A variety of bibliographic elements were collected, including annual publications, authors, countries/regions, journals, keywords, etc.

RESULTS

A total of 691 publications were included in this study. Professor Fabrice Bartolomei of Timone Hospital in France was the most productive and influential author in the field of SEEG, whereas the authors from the United States were the leaders in general. In addition, we found that deep learning and source localization in SEEG have been popular in recent years.

CONCLUSION

This study provides a comprehensive analysis of SEEG research and highlights the growing interest in SEEG and its deep learning and source localization.

Key Words: Stereoelectroencephalography; Neurology; Bibliometric analysis; Science Core Collection; VOSviewer; CiteSpace

Core Tip: Stereoelectroencephalography (SEEG) is a popular issue recently. The United States, China, and France are in the leading place. Epilepsia, Clinical Neurophysiology, Journal of Neurosurgery, and Frontiers in Neurology are the most popular journals in this field. Epilepsy is still the primary clinical application of SEEG, while deep learning and source localization have strongly increased.

INTRODUCTION

Stereoelectroencephalography (SEEG) was first formed in the 1950s at Saint Anne Hospital in France and was designed and developed in the 1960s in France by Talairach et al[1]. SEEG uses stereotactic techniques to place intracranial electrodes to record intracranial electrical activity and direct electrical stimulation to identify epileptogenic areas in patients with focal epilepsy[2-4]. SEEG is a diagnostic tool for drug-resistant focal epilepsies, showing precise positioning, minimal trauma, high patient tolerance, and other advantages[5,6]. A recent study demonstrated the performance of deep learning on SEEG signals and demonstrated that the deep learning method was beneficial for the decoding process because it is good at translating brain signals into commands for controlling external devices[7]. Another study showed that SEEG source localization is useful for corroborating the epileptogenic zone because it is inherently three-dimensional and can be modeled through source localization[8]. As a result, SEEG has attracted increasing attention in the research field of deep learning and source localization for drug-resistant focal epilepsies.

Bibliometrics is a technique employed to assess and examine the advancement of a specific field of science[9]. It utilizes computer technologies to present visual representations of literature analysis outcomes in a concise and comprehensible graphical format[10]. Overall, bibliometric analysis offers a comprehensive approach to evaluating academic productivity, summarizing academic frontiers and hotspots, and predicting trends within a particular research field. By understanding the results of bibliometric analysis, researchers can gain valuable insights into the scholarly landscape and make informed decisions regarding their research endeavors.

This study aims to analyze the bibliometric characteristics of the original studies on SEEG over the past 5 years; to analyze the publication trends, popular research topics, and cutting-edge trends in this field; and to provide references for clinicians and researchers.

MATERIALS AND METHODS

Data source and search strategy

The literature was retrieved from the Web of Science Core Collection (WoSCC) on 15 November 2023. After the inclusion and exclusion criteria were determined, the following formula was used as the retrieval strategy: TS = “stereotactic EEG” OR “stereoelectroencephalography” OR “SEEG” OR “stereoelectroencephalography SEEG” OR “intracranial EEG” OR “intracranial encephalography” OR “IEEG”. The literature type was restricted to “article”, and the publication year was limited to 2019 to 2023. To adequately retrieve the relevant literature, an exact search was not performed.

Inclusion and exclusion criteria

The original research related to SEEG, identified through the WoSCC database and published between 2019 and 2023, was further screened. Conversely, reviews, conference papers, letters, editorials, and notes, as well as retracted papers, were excluded from the analysis. There were no language restrictions.

Data screening and extraction

To ensure the reliability of the results, two independent researchers meticulously selected the literature by reading the abstracts and methods. Divergence in literature selection was settled by discussion or consultation with specialists. Consequently, all retrieved records were downloaded in “plain text” and “Excel” formats. Relevant parameters, such as titles, publication years, citation times, authors, countries/regions, institutions, journals, keywords, usage counts, WoS categories, etc., were extracted for further analysis.

Data analysis and visualization

Descriptive statistics from Microsoft Excel (V2021) were used to present the publication trends and cumulative citations. VOSviewer (V. 1.6.20) and CiteSpace (V. 6.2. R6) software were applied to perform knowledge mapping analysis, including the author, institution, country/region, journal, and keywords. Keywords serve as concise summaries of a paper’s core content and essence, and through keyword co-occurrence analysis, researchers can gain insights into the research hotspots within a scientific domain. In this study, VOSviewer was employed to construct a network visualization of keyword cooccurrences from the 691 documents. We merged synonyms such as “stereoelectroencephalography”, “stereotactic EEG”, “stereotactic electroencephalography”, “stereoelectroencephalography” and other synonyms into “SEEG”.

VOSviewer was employed to conduct coauthorship analyses of the author, institution country/region, and journal. VOSviewer also provided us with the publication and citation counts. The relatedness of the nodes in VOSviewer figures is determined by their number of coauthored publications. In addition, we applied VOSviewer to conduct a co-occurrence analysis for the keywords. Finally, CiteSpace was used to perform the citation burst analysis of keywords.

RESULTS

In the present study, we identified 6453 records by the above formula in the WoSCC on 15 November 2023. A total of 2339 publications were published between 2019 and 2023, of which 1831 records were tagged “Articles”. After further screening the document type and relevance, 691 original research publications were ultimately included. The number of articles written in English, German, and Russian are 689, 1, and 1, respectively. The study procedure is shown in Figure 1.

Figure 1 Flow chart of the main steps of this study.

Annual trend of publications

Our analysis revealed that the number of publications markedly increased from 2019 to 2021 (a peak level of 169 in 2021) and then decreased from 2022 to 2023 (Figure 2). The mean number of publications per year was 138.3, and the median was 143 (Figure 2). The cumulative number of citations of these articles increased from 1591 to 4246 during the last five years (Figure 2).

Figure 2 Annual trends of global scientific publications from 2019 to 2023.

The most productive and influential authors

The publication and citation numbers of the authors were identified via VOSviewer software. In total, 3161 authors participated in publishing the articles. The top 10 authors are listed in Table 1. Fabrice Bartolomei (62 publications) was far ahead of the others, emerging as the most prolific author. Other productive authors included Lagarde Stanislas (33 publications), Zhang Kai (25 publications), Carron Romain (24 publications), McGonigal Aileen (23 publications), Zhang Chao (23 publications), Wang Xiu (21 publications), etc. With 577 citations, Fabrice Bartolomei was also the most influential researcher in this field, followed by Lagarde Stanislas (387 citations) and McGonigal Aileen (324 citations).

Table 1 Top 10 authors in stereoelectroencephalography research.

Rank	Author	Documents	Citations
1	Fabrice Bartolomei	62	577
2	Lagarde Stanislas	33	387
3	Zhang Kai	25	181
4	Carron Romain	24	301
5	Mcgonigal Aileen	23	324
6	Zhang Chao	23	156
7	Wang Xiu	21	173
8	Trebuchon Agnes	20	278
9	Yu Tao	18	46
10	Hu Wenhan	15	91

Most productive and influential scientific organizations

VOSviewer software was employed to identify the publication and citation numbers of the scientific organizations. A total of 901 institutions published their works in the last five years. Among the 691 publications, 529 involved multiple institutions, and 196 represented international collaborations. Capital Medical University published 85 articles on SEEG, ranking first, followed by Aix-Marseille University (77 articles), Timone Hospital (48 articles), Fudan University (24 articles), and Shanghai Jiao Tong University (23 articles), and the top 10 organizations are listed in Table 2. Aix-Marseille University had the most citations (n = 701), followed by Timone Hospital (n = 462) and Capital Medical University (n = 378).

Table 2 Top 10 organizations in stereoelectroencephalography research.

Rank	Organizations	Count	Citations	Country
1	Capital Medical University	85	378	China
2	Aix-Marseille University	77	701	France
3	Timone Hospital	48	462	France
4	Fudan University	24	100	China
5	Shanghai Jiao Tong University	23	83	China
6	University College London	22	179	United Kingdom
7	University of Pittsburgh	22	86	United States
8	Grande Ospedale Metropolitano Niguarda	21	291	Italy
9	McGill university	20	194	Canada
10	Cleveland clinic	18	186	United States

Distribution analysis of the country and regions

Country analysis was performed by VOSviewer. Our evaluation revealed that the articles belonged to 45 countries or regions. The United States of America has the highest number of publications (n = 270) on SEEG, far ahead of China (n = 155) and France (n = 132), followed by Italy (n = 53) and Canada (n = 51). The rule of the citation rank was similar to that of the publication number. Figure 3 shows the citation relationships of the countries or regions that had at least 5 publications and Table 3 shows the top 10 countries in terms of the publications. Notably, numerous countries maintain strong connections with both the United States and France. Nevertheless, countries in yellow, including China, Spain, Austria, Belgium, Romania, and the Czech Republic, boomed in this field.

Figure 3 Citation analysis of the countries or regions. The threshold for each node presented in the map is set as 5 publications.

Table 3 Top 10 countries in stereoelectroencephalography research.

Rank	Country	Documents	Citations
1	United States	270	1778
2	China	155	658
3	France	132	1118
4	Italy	53	484
5	Canada	51	394
6	United Kingdom	48	411
7	Germany	29	224
8	Australia	25	153
9	Spain	22	117
10	Japan	18	119

Popular journals

We utilized VOSviewer to analyze the journals that published the articles. These articles were recorded by 156 journals, and 10 journals recorded more than 20 articles. Table 4 shows the top 10 journals in terms of the publications. With 45 documents, Epilepsia published the largest number of articles in this field, followed by Clinical Neurophysiology (39 documents) and the Journal of Neurosurgery (32 documents). In terms of the total number of citations, Epilepsia (n = 549), Journal of Neurosurgery (n = 243), and Clinical Neurophysiology (n = 224) also ranked in the top 3. However, yellow nodes in the citation analysis map (Figure 4), such as Epilepsia, Neurology, and Cortex, had higher average citations.

Figure 4 Citation analysis of the journals. The threshold for each node presented in the map is set as 5 publications.

Table 4 Top 10 journals of the number of publications on stereoelectroencephalography studies.

Rank	Journal	Documents	Citations
1	Epilepsia	45	549
2	Clinical Neurophysiology	39	224
3	Journal of Neurosurgery	32	243
4	Frontiers in Neurology	27	93
5	Epilepsy Research	25	117
6	Epilepsy & Behavior	24	163
7	Seizure	21	125
8	Journal of Neural Engineering	21	81
9	Epilepsy and Behavior Reports	21	31
10	Epileptic Disorders	21	19

Keywords analysis

The results are presented in Figure 5. Keywords with a frequency of 10 or higher were presented. As depicted, the identified keywords are organized into 5 distinct clusters. The green cluster primarily encompasses keywords pertinent to the clinical applications of SEEG, and the red cluster focuses on encephalic regions and functions.

Figure 5 Keywords network. The lowest frequency in the articles of these nodes is 10.

The related research on SEEG encompasses various aspects, such as its working principle, action site, and research object. By utilizing relevant keywords, ten distinct clusters were generated, providing a comprehensive overview of the field. Figure 6 depicts the temporal evolution of keywords via the utilization of CiteSpace software, revealing a dynamic research domain marked by evolving topics, including direct electrical stimulation, the epileptogenic zone, and medically refractory epilepsy. Notably, epilepsy was the primary topic of SEEG, other issues such as sleep and memory were also involved. By further analysis with CiteSpace, we obtained the keywords with the strongest citation bursts (Figure 7). Among them, deep learning and source localization have booming tendencies.

Figure 6 Timeline visualization of keyword analysis. SEEG: Stereoelectroencephalography.

Figure 7 The 10 keywords with the strongest citation bursts in the past 5 years.

DISCUSSION

Bibliometric analysis has emerged as an effective, objective, and reproducible tool for analyzing research output within a specific field[11]. In the present study, we analyzed 691 articles that investigated SEEG. Interestingly, the number of publications in this field remains high. This result highlights the great interest of global researchers in SEEG, underscores the value of SEEG, and highlights its importance across the board. It is therefore likely that the emphasis on SEEG will continue in the future.

In terms of author contributions, more than 3000 researchers have published related articles. Professor Fabrice Bartolomei, an expert on epilepsy from Timone Hospital in France, is the most productive and influential researcher in the field of SEEG. His first studies on epilepsy demonstrated the mechanism of seizure-related humming via SEEG techniques[12]. Many clinical investigations on epilepsy have been conducted since then. The most frequently cited study was published in 2019, indicating the determinants and prognosis of seizure onset patterns for focal epilepsies[13]. Recently, his team focused on investigating SEEG markers for epilepsy[14,15].

The United States, China, and France ranked in the top three countries in terms of the number of publications, which means that these three countries represent the highest level of SEEG research worldwide and make the greatest contribution to this field. The co-authorship map of countries revealed global cooperation in this field, with the United States and France exhibiting notably greater centrality. However, China, which ranks in the top three by publications, has relatively less cooperation with other countries. Therefore, China should enhance its international cooperation and carry out more comprehensive and profound research on SEEG.

The present study revealed that the most highly cited journal in this field was Epilepsia, with 45 documents and 549 citations, followed by Clinical Neurophysiology, the Journal of Neurosurgery, and Frontiers in Neurology. Therefore, we believe that these journals will continue publishing influential and advanced research in this field in the future. Experts and scholars who have a keen interest in the subject will discover a plethora of novel developments and cutting-edge research in these journals.

Keywords encapsulate the core of papers, revealing the hotspots of research. Keyword co-occurrence analysis pinpoints scientific hotspots. Visual network diagrams clarify evolving trends. Burst keywords track historical development and suggest future research directions. The analyses of keywords revealed that some keywords, such as “surgical technique”, “effective connectivity” and “seizure onset patterns”, have been researching hotspots in this field in the past. SEEG’s accurate identification determines the surgical prognosis, and this characteristic has shifted the research trend toward ‘surgical technique’. However, “deep learning” and “source localization” have been the most popular topics in recent years. Clinical research can focus on these aspects. Machine learning is playing an increasingly important role in medical image analysis. In contrast to these traditional decoding techniques, the equivalent or superior performance of deep learning methods has been demonstrated across various studies utilizing SEEG[16,17]. Ictal SEEG source localization is valid and useful for corroborating the epileptogenic zone[18]. Therefore, it is important to develop deep learning and source localization for SEEG, and future research on deep learning and source localization will involve more clinical applications.

Currently, our knowledge regarding epilepsy networks and human brain functional networks remains limited. However, SEEG plays an important role and has emerged as the preferred method for intracranial monitoring in epilepsy patients[4]. As the diagnostic and therapeutic approaches based on SEEG continue to evolve, they promise to significantly increase the quality of life for individuals living with drug-resistant epilepsy[19,20]. Based on these analyses, we predict that these studies on deep learning and source localization will continue to be hotspots in the field of SEEG.

Limitations

This study has certain limitations. First, the exclusive reliance on data from the WoSCC database presents a challenge, as it inevitably leads to concerns regarding the comprehensiveness of the analyzed data. Nevertheless, the WoSCC database encompasses a vast repository of high-quality academic journals, conference papers, and patents from across the globe. It is the articles of superior quality, rather than those of subpar quality, that catalyze the advancement of scientific research. Second, research results generally reach their peak of citations one to two years after publication or release, which can lead to a lag in the evaluation results of bibliometric analysis and may not reflect the latest research results and trends over time.

CONCLUSION

In this study, 691 articles published in the last 5 years were analyzed via bibliometrics to clarify the current status and future trends of SEEG research. The number of publications markedly increased from 2019 to 2023. Fabrice Bartolomei is the most prolific and influential author. The Capital Medical University of China ranks first in terms of publication number. Additionally, the United States has the highest number of publications and citations. Epilepsia, Clinical Neurophysiology, Journal of Neurosurgery, and Frontiers in Neurology are the most popular journals in the field of SEEG. Epilepsy is still the primary clinical application of SEEG, while deep learning and source localization have strongly increased recently. Overall, our study provides a useful reference for further research on SEEG.