Near-infrared spectroscopy in schizophrenia: A bibliometric perspective

Abstract

BACKGROUND

Compared with current methods used to assess schizophrenia, near-infrared spectroscopy (NIRS) has the advantages of providing noninvasive and real-time monitoring of functional activities of the brain and providing direct and objective assessment information.

AIM

To explore the research field of NIRS in schizophrenia from the perspective of bibliometrics.

METHODS

The Web of Science Core Collection was used as the search tool, and the last search date was April 21, 2024. Bibliometric indicators, such as the numbers of publications and citations, were recorded. Bibliometrix and VOS viewer were used for visualization analysis.

RESULTS

A total of 355 articles from 105 journals were included in the analysis. The overall trend of the number of research publications increased. Schizophrenia Research was identified as an influential journal in the field. Kasai K was one of the most influential and productive authors in this area of research. The University of Tokyo and Japan had the highest scientific output for an institution and a country, respectively. The top ten keywords were “schizophrenia”, “activation”, “near-infrared spectroscopy”, “verbal fluency task”, “cortex”, “brain, performance”, “working-memory”, “brain activation”, and “prefrontal cortex”.

CONCLUSION

Our study reveals the evolution of knowledge and emerging trends in the field of NIRS in schizophrenia. the research focus is shifting from underlying disease characteristics to more in-depth studies of brain function and physiological mechanisms.

Key Words: Near-infrared spectroscopy; Schizophrenia; Bibliometrics; Bibliometrix; VOS viewer

Core Tip: The assessment scale is often used to evaluate patients with schizophrenia in clinical practice, but there are still some limitations of the scale due to its dependence on subjective judgment and its inability to directly observe brain function, as well as the difficulty in accounting for individual differences among patients. Near-infrared spectroscopy (NIRS) has been used with increasing frequency in clinical settings over the past 20 years. However, there is a lack of bibliometric analysis of NIRS in schizophrenia. Our study demonstrates the evolution of knowledge and emerging trends in the field of NIRS for patients with schizophrenia through the use of bibliometric methods and visualization tools.

INTRODUCTION

Schizophrenia is a severe, chronic, disabling mental illness that typically manifests in late adolescence or early adulthood[1]. Moreover, the course of schizophrenia is highly heterogeneous; that is, different patients have different symptoms, forms of onset, and treatment responses. During the treatment of schizophrenia, patients' symptoms may fluctuate and repeat, which increases the degree of uncertainty in the prognosis. At present, the assessment scale is often used in clinical practice to evaluate patients with schizophrenia, but there are still some limitations of the scale due to its dependence on subjective judgment and inability to directly observe brain function as well as the difficulty in accounting for individual differences among patients.

Compared with current schizophrenia assessment methods, many noninvasive testing devices have the advantage of real-time monitoring of functional activities of the brain, providing more direct and objective assessment information. For example, functional magnetic resonance imaging can indirectly reflect functional activities of the brain by measuring the contrast of blood oxygenation level-dependent signals[2]. In addition, electroencephalography has high temporal resolution and is able to capture rapid changes in brain activity[3]. However, near-infrared spectroscopy (NIRS) has significant advantages over these noninvasive testing devices in the diagnosis of patients with psychiatric disorders. The equipment used for NIRS is lightweight, easy to use in real-time monitoring, and insensitive to motion artifacts, which reduce patient anxiety and improve patient acceptance. Therefore, NIRS is more suitable for psychiatric studies that require assessment of brain activity in a clinical setting[4].

In fact, NIRS has been one of the most quickly adopted analytical techniques in recent decades[5]. NIRS is a vibration spectrum technology that has several applications in biological research and biomedical diagnostics, including blood and bodily fluid analysis and cell and biomolecular research[6]. In the medical field, it can be utilized to predict the success rate of extubation during newborn cardiac surgery in addition to being a predictor of intraoperative cerebral ischemia[7,8]. In 2004, Suto discovered that NIRS, with its noninvasive nature and high temporal resolution, can be a useful tool for psychiatric research[9]. To date, NIRS has been employed to assess alterations in blood flow and brain activity in individuals suffering from mental illnesses, including schizophrenia and depression[10-12]. For example, a meta-analysis revealed that NIRS was able to detect under activation in the frontotemporal region of patients with major depression and schizophrenia, and their topographic distribution was disease-specific[13].

Bibliometric analysis is a quantitative approach to researching and assessing articles. It can be used to evaluate current research findings efficiently, find important papers that offer insightful analysis and solid proof, and help determine the best course for future research[14,15]. At present, Bibliometrix, Cite space, and VOS viewer are the main tools used for bibliometric analysis and visualization[16-18]. Compared with Cite Space, VOS viewer is far easier to use and is capable of fulfilling all of the requirements for core bibliometric analysis, such as identifying institutional distributions, tracking recent advancements, analyzing major researchers and influential literature, and tracking overall research trends within related fields. Importantly, a powerful visualization package called Bibliometrix was developed. It can automatically process large amounts of data and generate various scientific maps and statistical indices. When used in tandem, Bibliometrix can supplement the outputs from VOS viewer to enable a more thorough visual bibliometric analysis by providing more data and charts that.

NIRS has been used with increasing frequency in clinical settings over the past 20 years, and previous studies have emphasized the importance of NIRS in the diagnosis of diseases, especially in the fields of neurology and psychiatry[13,19-21]. However, there are few bibliometric studies on NIRS in schizophrenia, and we cannot understand the trends and characteristics of the field without such analysis. Consequently, through the use of bibliometrics and visualization tools, we systematically explored the how the application of NIRS developed for patients with schizophrenia, with the goal of identifying research hotspots and frontiers and providing valuable references and guidance for researchers in related fields.

MATERIALS AND METHODS

Search tool and search strategy

The Web of Science Core Collection (WoSCC, index: Science Citation Index Expanded) was used as the search tool, and the search date was April 21, 2024. We used “TS = (“near-infrared spectro” OR “near-infrared spectroscopy” OR “near infrared spectro” OR “near infrared spectroscopy” OR “NIR spectro” OR “NIR spectroscopy” OR “NIRS”) AND TS = (“schizophreni” OR “schizophrenic disorde” OR “schizophrenic disorder”)” as the search strategy. Two researchers carried out the searches in parallel.

Data collection and analysis

The R-based Bibliometrix package (version 4.3.1) and VOS viewer (version 1.6.18) were used for visualization analysis. Microsoft Excel 2021 (2403 Build 16.0.17425.20176) was used as the table processing software.

First, the literature data were exported from the WoSCC and imported into the Bibliometrix software package, and the overview function of the software was used to obtain the main information, annual scientific research output, and annual average citations. Second, a source analysis tool was used to collect the output and total citations of the journals. The Author Analysis module was used to analyze outputs, citations, and collaborations by authors, countries, and institutions. Third, word frequency and keyword cloud maps were obtained via keyword analysis. Finally, the evolution function of theme words was used to produce the evolution of keywords over two periods. Then, VOS viewer was used to create a partnership map for authors, countries, and institutions and a keyword co-occurrence map.

RESULTS

Overview

A total of 355 articles from 105 journals were included in the analysis. The articles were published from 1994 to 2024. The annual growth rate reported in the literature was 6.7%. On average, 6 authors contributed to each article. Every article received an average of 23.13 citations. Figure 1 shows the proportions of article types; 277 articles, 36 meeting abstracts, 26 reviews, and 16 other types were included.

Figure 1 Proportions of article types.

Production

Figure 2A depicts the change in annual production trends from 1994 to the present. Between 1994 and 2014, the number of publications fluctuated; the number of publications decreased to 12 in 2019, rose to the highest point in 2021 (28 publications), and then fell again. In general, the overall trend was upward, but there were intermediary times where there was a decline. In terms of citations, Figure 2B shows that the annual average number of citations fluctuated.

Figure 2 Publications and citations in the field of near-infrared spectroscopy in schizophrenia. A: Annual number of publications; B: Average number of citations per year.

Journals

The top ten relevant journals are shown in Table 1. The top three in terms of publication number were Schizophrenia Research, which published 29 articles; Frontiers in Psychiatry, which published 25 articles; and Scientific Reports, which published 21 articles. The top three journals accounted for 20.8% of the total number of publications. In terms of citations, Schizophrenia Research was cited 1002 times, ranking first among journals. Biological Psychiatry came in second with 503 citations. Scientific Reports was the third most cited journal, with 406. These data reveal the ample advantage that Schizophrenia Research holds, revealing it to be an authoritative journal of NIRS in schizophrenia.

Table 1 Top 10 relevant journals.

Journal	Number	Citation	IF	JCR
Schizophrenia Research	29	1002	4.5	Q2
Frontiers in Psychiatry	24	233	4.7	Q2
Scientific Reports	21	406	4.6	Q2
Biological Psychiatry	17	503	10.6	Q1
Journal of Psychiatric Research	16	394	4.8	Q2
Journal of Affective Disorders	16	339	6.6	Q1
Psychiatry Research-Neuroimaging	14	297	2.3	Q3
Progress in Neuro-Psychopharmacology & Biological Psychiatry	13	345	5.6	Q1
Psychiatry and Clinical Neurosciences	13	241	11.9	Q1
Plos One	10	233	3.7	Q2

IF: Impact factor; JCR: Journal citation reports.

Authors, institutions, and countries/regions

Table 2 shows the H-index, citations, and other indicators of the top 10 authors. The most influential author was Kasai K., who was followed by Takizawa R, Nishimura Y. and Pu S. In terms of citations, Kasai K ranked first with 1171 citations, followed by Fukuda M and Takizawa R; their numbers of citations were 1060 and 1046, respectively. The H-index is a hybrid quantitative indicator that can be used to evaluate the amount of academic output of a researcher or journal relative to the level of academic output[22]. Kasai K has the highest, with an H-index of 19, followed by Takizawa R, with an H-index of 18. Moreover, there was close cooperation between the authors. Kasai K who is from Japan, published many articles with and worked closely with other authors, as shown in Figure 3A. Chou PH, Pu S, and Marumo K also worked closely with the other authors.

Figure 3 Collaborative networks in the field of near-infrared spectroscopy in schizophrenia. A: Map of collaboration among authors; B: Map of collaboration among Institutions; C: Map of collaboration among countries/regions.

Table 2 The top 10 most influential authors.

Element	H-index	Number	Total citation	Country	Institution
Kasai K	19	41	1171	Japan	University of Tokyo
Takizawa R	18	34	1046	Japan	University of Tokyo
Nishimura Y	13	25	482	Japan	University of Tokyo
Pu S	13	19	674	Japan	Tottori University
Fallgatter AJ	11	15	487	Germany	University of Tubingen
Fukuda M	11	18	1060	Japan	Saga University
Nakagome K	11	17	445	Japan	National Center of Neurology and Psychiatry
Chou PH	10	17	223	China	China Medical University
Kaneko K	10	14	336	Japan	Tottori University
Koike S	10	22	345	Japan	University of Tokyo

Table 3 shows the top ten most relevant institutions. The University of Tokyo ranked first, with 129 articles; Tottori University had 41 publications, and Osaka University had 35 publications. The cooperation among institutions is shown in Figure 3B. The University of Tokyo published the most papers and had the most collaborations with other institutions. Taichung Veterans General Hospital, the National Center of Neurology and Psychiatry (Japan), and Tottori University collaborated with the University of Tokyo.

Table 3 The top 10 institutions.

Institution	Number
University of Tokyo	129
Tottori University	41
Osaka University	35
Tsinghua University	30
Taichung Veterans General Hospital	27
Nara Medical University	25
Ankara University	24
Chiba University	24
Peking University	24
Fukushima Medical University	22

A total of 20 countries had relevant studies, and the top 10 are shown in Table 4. Japan ranked first, with 190 articles published. China ranked second with 68 articles, and Germany (n = 25) ranked third. In terms of article citations, the top three are Japan, Germany, and the United States; their published articles were cited 4790 times, 936 times, and 860 times, respectively. When ranked according to the average number of article citations, the United States ranks first. The average number of citations is 57.3. In second place was Australia (n = 54.0), which published 2 articles and obtained 108 citations. In addition, Singapore was in third place, with an average of 53.2 citations. Figure 3C shows country cooperation. The countries where researchers worked closely with those from other countries were Japan, the People's Republic of China, and the United States. In addition, Germany, Italy, Turkey, Australia, Singapore, and Canada cooperate internationally.

Table 4 Top 10 most influential countries.

Country	Number	SCP	MCP	Citation	Average citation
Japan	190	173	17	4709	24.8
China	68	53	15	736	10.8
Germany	25	20	5	936	37.4
USA	15	9	6	860	57.3
Turkey	11	8	3	69	6.3
Italy	7	6	1	69	9.9
Singapore	6	2	4	319	53.2
India	5	5	0	70	14.0
Canada	4	1	3	106	26.5
Korea	3	2	1	133	44.3

SCP: Single-country authors; MCP: Multi-country authors.

Table 3 shows the top ten most relevant institutions. The University of Tokyo ranked first, with 129 articles; Tottori University had 41 publications, and Osaka University had 35 publications. The cooperation among institutions is shown in Figure 3B. The University of Tokyo published the most papers and had the most collaborations with other institutions. Taichung Veterans General Hospital, the National Center of Neurology and Psychiatry (Japan), and Tottori University collaborated with the University of Tokyo.

A total of 20 countries had relevant studies, and the top 10 are shown in Table 4. Japan ranked first, with 190 articles published. China ranked second with 68 articles, and Germany (n = 25) ranked third. In terms of article citations, the top three are Japan, Germany, and the United States; their published articles were cited 4790 times, 936 times, and 860 times, respectively. When ranked according to the average number of article citations, the United States ranks first. The average number of citations is 57.3. In second place was Australia (n = 54.0), which published 2 articles and obtained 108 citations. In addition, Singapore was in third place, with an average of 53.2 citations. Figure 3C shows country cooperation. The countries where researchers worked closely with those from other countries were Japan, the People's Republic of China, and the United States. In addition, Germany, Italy, Turkey, Australia, Singapore, and Canada cooperate internationally.

Keywords

Figure 4 shows the co-occurrence of keywords. The top ten keywords were “schizophrenia”, “activation”, “near-infrared spectroscopy”, “verbal fluency task”, “cortex”, “brain performance”, “working-memory”, “brain activation”, and “prefrontal cortex”. The popular topics are shown in Figure 5A. The most recent hot topics were “bipolar”, “unipolar”, and “high risk”. The evolution of keywords is shown in Figure 5B. The evolution of keywords was roughly divided into two stages: From 1994 to 2016 and from 2017 to 2024. Some keywords were merged, and some keywords were derived from new keywords.

Figure 4 Map of keyword co-occurrence.

Figure 5 Keyword analysis. A: Keyword fashion trend; B: Keyword evolution.

DISCUSSION

Our study focused on the current status and trends of NIRS in schizophrenia. In recent years, the number of publications on the clinical use of NIRS in schizophrenia has generally increased. This increase may be attributed to the prominent advantages of NIRS in clinical practice, such as its noninvasiveness, real-time data acquisition, and high spatiotemporal resolution, which has led to its broad application prospects in schizophrenia diagnosis, efficacy evaluation, and rehabilitation training[4,23]. Moreover, with the continuous improvement of technology and the reduction in cost, NIRS technology has gradually been accepted and applied by more medical institutions and research teams[9,24]. These advances have promoted the expansion of research in this field and provided new ideas and methods for the clinical diagnosis and treatment of schizophrenia.

The rising popularity and academic exchange activity of research on NIRS in schizophrenia

Compared with bibliometric studies in other fields, the number of articles published in meetings in the field of NIRS in schizophrenia is relatively high[25,26]. Meeting articles are important channels through which the academic community can exchange research results; these channels promote academic cooperation and knowledge sharing in the field and promotes the expansion of research, and they also strengthen mutual inspiration between peers, thus helping to improve the research impact and influence of the whole field[27]. This means that there is a high degree of research activity in the field of NIRS in schizophrenia, researchers are actively engaged in studying this cutting-edge technology, and the results have been widely communicated and recognized.

Key journals driving research into NIRS in Schizophrenia

The distribution of journals in this field exhibits a notable concentration trend, with nearly half of all the publications published in these journals. Schizophrenia Research stands out as an authoritative journal due to its substantial number of published articles and remarkable citation numbers, reflecting its strong influence and academic value. Additionally, journals such as Frontiers in Psychiatry and Scientific Reports have gained wide recognition because of their stable publication rates and impressive citation counts. These journals are distributed in the Q1 and Q2 zones of the JCR, indicating their high quality and rigorous selection criteria. Overall, these journals provide high-quality publishing platforms for researchers in the field while also reflecting the focus of research and academic concerns in the field. Understanding the distribution and influence of these journals is beneficial for researchers in choosing the appropriate journal for submitting their work and enhancing the dissemination and impact of their research outcomes.

Collaborative networks as pivotal catalysts for advancing research on NIRS in schizophrenia

Analysis of authors can reveal not only the main researchers in a certain field and their research contributions but also the cooperative relationships and networks among researchers; furthermore, it can reveal the academic exchanges and interactions between individuals[28]. Kasai K and Takizawa R are similarly highly ranked in terms of the H-index, which means that their academic output is not only high in quantity but also high in quality; each article has been cited a fairly high number of times, showing that they are highly influential authors in the field. Notably, these authors are not isolated researchers but are close collaborators. For example, Kasai K is from Japan and has not only published many articles but also carried out in-depth collaborative research with other authors, which helps promote academic exchange and resource sharing, and it promotes the thorough advancement of research.

In terms of institution analysis, the University of Tokyo has performed outstanding research in the field of NIRS in schizophrenia, not only in terms of the number of publications but also in terms of extensive cooperation. Tottori University and Osaka University are also actively involved in research and promote the development of this field through cooperation. With respect to country/region analysis, Japan holds a leading position in research in the field of NIRS in schizophrenia, topping the list with 190 publications. China ranked second with 68 publications, showing the country’s rapid development and momentum that is helping their researchers to catch up in the field. China and Japan account for more than half of the total number of publications, indicating that research in this field is more popular in Asia than in other parts of the world. However, when ranked according to the average number citations per article, the United States ranks first, which indicates that the quality of its research results and academic influence are greater. Moreover, international cooperation plays an important role in the field of NIRS in schizophrenia. Japan, China, and the United States are the countries that cooperate the most closely with other countries, and they have established international cooperation relationships with several countries, including Germany, Italy, and Turkey. This further demonstrates that high academic productivity, high-quality research, and high impact cannot be achieved without international cooperation, which enables researchers to share resources and jointly solve research problems.

Current hotspots and future directions

Keyword analysis not only reveals the hotspots of current research but also provides strong guidance for future research directions. With respect to the co-occurrence of keywords, the top ten keywords are related mainly to schizophrenia, NIRS, and the application of NIRS to the study of brain activation and cortical function. This suggests that research in this area is focused on the use of NIRS to study brain activation and cortical functional performance in patients with schizophrenia while performing different tasks, such as verbal fluency tasks and working memory tasks[29-31]. Of note, researchers are also beginning to look at changes in brain function in people with psychiatric disorders other than schizophrenia and those at high risk, meaning that future research might focus on biphasic, unipolar, and high-risk patients, possibly because the pathogenesis of these diseases is similar. Moreover, these research results can help to further expand the range of clinical indications for NIRS and aid in the differential diagnosis of schizophrenia[32-34].

In addition, looking at the evolution of keywords, research in this field can be roughly divided into two stages. From 1994 to 2016, research focused more on the early identification of diseases, clinical manifestations, and exploration of pathological mechanisms. With the continuous progress of technology and depth of research, a relatively stable research framework and keyword system have gradually formed. From 2017 to 2024, there was a clear shift in research hotspots and frontiers. The new keywords that started to dominate this field included “prefrontal cortex”, “metabolism”, “memory”, “sleep”, “experience”, and “functional connectivity”. These changes reflect a shift in research focus from underlying disease characteristics to more in-depth studies of brain function and physiological mechanisms. Specifically, the prefrontal cortex, an important region of the brain, plays a key role in the pathogenesis of schizophrenia. Researchers are beginning to pay more attention to functional abnormalities in the prefrontal cortex and their associations with schizophrenia[35,36]. Moreover, physiological factors such as metabolism and sleep are also considered important factors affecting the onset and course of schizophrenia, and changes in memory and experience are also common symptoms of schizophrenia patients; therefore, these aspects have gradually become a research hotspot in this field[37,38].

CONCLUSION

Our study demonstrates the evolution of knowledge and emerging trends in the field of NIRS for schizophrenia through the use of bibliometric methods and visualization tools. Schizophrenia Research is an influential journal in the field. Kasai K is one of the most influential and productive authors in this area of research. The University of Tokyo and Japan had the highest scientific output for an institution and a country, respectively. The United States had a higher quality of research and academic influence in this field than other countries. Additionally, high academic productivity, high-quality research, and high impact cannot be achieved without international cooperation. Finally, research in this field can be roughly divided into two stages. In the early stage, researchers focused more on the early identification of diseases, clinical manifestations, and exploration of pathological mechanisms. At present, the research focus is shifting from underlying disease characteristics to more in-depth studies of brain function and physiological mechanisms.

This study has several limitations. Bibliometric analysis relies mainly on published scientific research papers, and visualization tools rely on published literature data. The field of NIRS in schizophrenia is characterized by complex and diverse research, so there may be many valuable but unpublished research studies; the absence of these data may lead to bias. Moreover, bibliometric analysis focuses mainly on quantitative indicators, such as the number of publications and citation frequency. Although quantitative indicators can reflect the activity and influence of the research field, they cannot fully represent the quality of the research[39]. In addition, research hotspots within a particular field change over time, which will require real-time updates.