1 Introduction

As a green and effective therapeutic method, acupuncture has a history of over 3,000 years in China and is widely applied in the treatment of neurological, endocrine, and respiratory systems (Wen et al., 2021). Despite its study since the 18th century and significant progress in clinical practice as well as in the understanding of acupoints and meridians, the clinical efficacy and physiological and biological mechanisms of acupuncture remain incompletely elucidated (Zhuang et al., 2013). Functional magnetic resonance imaging (fMRI), a leading non-invasive neuroimaging technique, offers excellent spatial and temporal resolution. Since the 1990s, fMRI has been applied in acupuncture research, including studies on the characterization of needling stimulation and brain responses before and after acupuncture (Napadow et al., 2008). Currently, fMRI is frequently used to evaluate the clinical effects of acupuncture interventions or to investigate its mechanisms of influence on the central nervous system (Yan et al., 2022).

This study employs Citespace and VOSviewer for visualization analysis. Citespace is used to analyze annual publications, countries, institutions, keywords, and co-cited references, while VOSviewer is applied to examine authors and co-cited authors, journals, and co-cited journals. The aim is to explore the application of fMRI in the field of acupuncture and conduct an in-depth analysis of its current research status and trends, thereby providing guidance and a foundation for further studies on the application of fMRI in acupuncture.

2 Materials and methods 2.1 Data source and retrieval strategy

The literature utilized in this study was sourced from the SCI-EXPANDED database within the Web of Science Core Collection, which is accessible at Changchun University of Traditional Chinese Medicine Library. The retrieval period spanned from January 1, 2004, to April 30, 2024. The retrieval process was concluded on May 26, 2024. The database search strategy encompassed the utilization of keywords “acupuncture” and “fMRI.”

2.2 Criteria for inclusion and exclusion of literature

Inclusion criteria: article or review article. The language is English, without any restrictions on the country of publication.

Exclusion criteria: editorial material, meeting abstract, proceeding paper, letter, correction, retracted publication, early access, book chapters, and literature unrelated to the topic.

2.3 Data collection and analysis

The data should be converted to “txt” format, named “download_*.txt,” and subsequently imported into Citespace (version 6.3 R1) and VOSviewer (version 1.6.20) for conducting data analysis on selected literature. The analysis includes information such as annual publications, journals, countries, institutions, authors, keywords, authors and their Co-citations, journal and their Co-citations, and references Co-citations. In CiteSpace, the parameter settings are as follows: (1) The time range is set from 2004 to 2024 with a yearly interval; (2) Only one node type is selected at a time; (3) The g-index selection k-value is set to 25; (4) The Top-N option is set to 50. In the Citespace network knowledge map, nodes represent the informational units of the analyzed items, with node size indicating the frequency of occurrence. The color and width of the “annual rings” within nodes reflect the timing and quantity of occurrences for the analyzed items, while a purple outer ring signifies high betweenness centrality within the overall network. The thickness of the connecting lines illustrates the strength of the relationships between different analyzed items.

3 Results 3.1 Document screening situation

We acquired a total of 967 sources, consisting of 838 articles, 103 review articles, 11 editorial materials, seven meeting abstracts, seven proceeding papers, five letters, and three corrections. Additionally, there were also three retracted publications and two early-access manuscripts. Furthermore, we identified one book chapter in our search. In terms of language, there are 964 articles written in English, one in Chinese, one in German, and one in Portuguese. After applying inclusion criteria, we selected a total of 937 articles and excluded an additional 380 articles that were deemed irrelevant to the topic. Therefore, a total of 557 records were included for further visualization and analysis in this study. Quantitative methods were employed to analyze the current status and trends of functional magnetic resonance imaging technology in acupuncture, following the process depicted in Figure 1. The task of conducting database searches and screening literature was assigned to two researchers, Zhongke Wang, and Lu Chen, while the responsibility of handling potential disputes or disagreements were undertaken by senior researcher Fuchun Wang.

Figure 1. Flow chart of literature screening.

3.2 Analysis of annual publications

The number of publications per year is shown in Figure 2, which can be divided into three stages. The first stage, from 2004 to 2010, saw slight fluctuations in the number of publications but exhibited a continuous growth trend, with a peak of 24 articles in 2009. However, the number of publications decreased to 15 articles in 2010. The second stage, from 2010 to 2017, was characterized by a steady increase in publication volume, although the rate of increase was relatively slow, with a decline to 22 articles in 2017. The third stage, from 2017 to 2024, witnessed a gradual increase in publications, particularly between 2019 and 2021, where the growth rate was notably higher, peaking in 2021. Based on data available up to April 2024, it is expected that the number of publications will continue to remain relatively high in the future.

Figure 2. The number of fMRI in the field of acupuncture for annual publications.

3.3 Analysis of national cooperative networks

The “Country” node should be selected for conducting a visual analysis of the countries mentioned in the literature using a knowledge graph. The study yielded a total of 23 nodes and 44 links, spanning across 23 countries or regions. The top five countries in terms of national publications are Peoples R China (422 articles), USA (116 articles), South Korea (62 articles), Germany (22 articles), and Taiwan (China) (16 articles). According to the centrality ranking, the top five countries are Peoples R China (0.79), USA (0.31), the Republic of Korea (0.2), Australia (0.16), and Switzerland (0.16, Table 1). The rankings of Peoples R China, USA, and South Korea consistently occupy the top three positions based on considerations of publication frequency and centrality. As shown in Figure 3A, the formation of China, USA, and South Korea as three cores signifies the presence of strong internal connections among these countries. The density value of 0.1739 suggests a relatively moderate level of inter-country cooperation.

Table 1. Top 10 authors related to fMRI in the field of acupuncture.

Figure 3. National cooperative networks (A) and institutional cooperative networks (B).

3.4 Network analysis of institutional collaboration

The “Institution” node should be selected to visualize and analyze the knowledge graph of institutions mentioned in the literature. The network consists of 305 nodes and 1,104 links, encompassing a total of 305 distinct institutions. Table 2 lists the top five institutions in terms of publication and centrality. Chinese Academy of Sciences takes the lead with the highest research output (74 articles), followed by Harvard University (65 articles) and Chengdu University of Traditional Chinese Medicine (65 articles), which are tied for second place. Massachusetts General Hospital ranks fourth (60 articles), while Beijing University of Chinese Medicine and Xidian University share fifth place with 58 articles each. In terms of centrality, the institutions are ranked as follows: Kyung Hee University (0.35), Harvard University (0.23), Chengdu University of Traditional Chinese Medicine (0.21), University of California System (0.2), and Massachusetts General Hospital (0.19, Table 3). The nodes of the Beijing University of Chinese Medicine, Chinese Academy of Sciences, Chengdu University of Traditional Chinese Medicine, and other institutions are prominently large in Figure 3B, indicating their central position and strong connections within their respective organizations. However, the density value is 0.0238, indicating a lack of cohesive collaboration among various institutions. Moreover, China boasts a substantial number of institutions engaged in fMRI research applied to acupuncture.

Table 2. Top five institution by publications.

Table 3. Top five institution by centrality.

3.5 Analysis of the authors and co-cited authors

The author collaboration network was generated using VOSviewer, which includes 2,706 authors who published 557 articles. In Figure 4A, authors with more than 7 publications are displayed. Each node represents an author, with the size of the node positively correlated with the number of publications. The connections between nodes represent collaborative relationships, with thicker lines indicating stronger cooperation. In the context of fMRI research related to acupuncture, the top 10 authors are shown in Table 4. Tian, J ranks first (n = 36), followed by Qin, W; Zeng, F; Bai, LJ; Kong, J; and Liang, FR. Among them, Kong, J and Liang, FR are tied for fifth place (n = 26). Analysis of the author collaboration network using VOSviewer reveals the formation of 8 clusters, with each cluster consisting of multiple authors engaged in long-term collaboration.

Figure 4. Map of the authors (A) and co-cited authors (B).

Table 4. Top five authors and co-cited authors.

The total number of co-cited authors is 10,734, among whom 12 authors have received more than 100 citations. Figure 4B illustrates the authors with citation frequencies exceeding 30, while Table 4 showcases the top five most frequently cited authors. Among them, Kong, J has received the highest number of citations with a count of 360. Following closely are Hui, KKS with 350 citations, Napadow, V with 259 citations, Bai, LJ with 241 citations, and Dhond, RP with 162 citations.

3.6 Analysis of journals and co-cited journals

Journal and journal Co-citation analysis was conducted using VOSviewer. Publications on the application of fMRI in the field of acupuncture were published in over 138 journals, with Figure 5A displaying the journals that appeared more than four times. Table 5 presents the top 10 most popular academic journals in the field, with Evidence-based Complementary and Alternative Medicine having the highest number of publications (n = 58), followed by Frontiers in Neuroscience (n = 30), Trials (n = 25), Frontiers in Neurology (n = 24), and Neural Regeneration Research (n = 20). Furthermore, Neuroscience Letters and Plos One are tied for tenth place. According to Journal Citation Reports 2023, Neural Regeneration Research has achieved a high impact factor (IF =5.9). Unfortunately, the majority of journals have an impact factor below five, and JCR categories are primarily concentrated within Q3.

Figure 5. Map of journals (A) and co-cited journals (B).

Table 5. Top 10 journal and co-cited journal.

The total number of cited journals is 3,278, out of which 6 journals have citation counts exceeding 500. Figure 5B shows that journals are cited more than 40 times. As shown in Table 5, the top 10 academic journals are Neuroimage (Co-citation = 1,618), followed by Hum Brain Mapp (Co-citation = 885), Pain (Co-citation = 766), Evidence-based Complementary and Alternative Medicine (Co-citation = 580), Plos One (Co-citation = 519) and Neuroscience letters (Co-citation = 502). Additionally, Proceedings of The National Academy of Sciences of The United States of America has the highest impact factor (IF = 9.4), followed by pain (IF = 5.9) and Neuroimage (IF = 4.7), while the remaining journals have impact factors below 5.

3.7 Analysis of co-cited references

A Co-citation analysis of references was conducted using the “Reference” feature, as shown in Figure 6A. The analysis revealed a network consisting of 825 nodes and 3,014 links, with a network density of 0.0089, indicating relatively weak connections between these 825 cited references. According to Table 6, the top five most frequently cited references include Dhond RP, 2008; Fang JL, 2009; Bai LJ, 2009; Hui KKS, 2005; and Bai LJ, 2009, which are considered representative. In terms of the centrality ranking of cited references, the top five are You YB, 2013 (0.21); Wang XY, 2016 (0.13); Chae Y, 2013 (0.11); Chen XY, 2015 (0.11); Kong J, 2009 (0.1); and Li ZJ, 2016 (0.1), as shown in Table 7. Notably, Kong J, 2009, and Li ZJ, 2016 (0.1) are tied for fifth place. It is worth mentioning that despite these two references being published later, Kong J and Li ZJ are highly representative of the application of fMRI in the field of acupuncture.

Figure 6. Map of co-cited references (A) and Top 25 keywords with the strongest citation bursts (B).

Table 6. Top five cited reference by frequency.

Table 7. Top five cited reference by centrality.

3.8 Co-occurrence analysis of keywords

The visualization analysis graph of the keyword co-occurrence network in this study consisted of 471 nodes and 2,667 links, resulting in a network density of 0.0241 (Figure 7A). The top 10 keywords, listed in descending order, include fMRI (223), stimulation (126), electroacupuncture (104), activation (104), pain (102), functional connectivity (101), connectivity (78), acupuncture (77), brain (76) and cortex (Zhang et al., 2021) as shown in Table 8. The frequency analysis indicates that fMRI is currently a research hotspot, with acupuncture having the highest centrality (0.24). A centrality value >0.1 suggests that this node has the highest attention and influence within the research field, representing a key area of focus in the domain.

Figure 7. Map of co-occurring keywords (A) and keyword clustering (B).

Table 8. Top 10 keywords.

3.9 Analysis of keyword clustering

The keywords about acupuncture research utilizing fMRI were examined through the implementation of the “K” clustering-LLR algorithm. Figure 7B results demonstrated that the modularity Q value surpassed the threshold of 0.3, reaching 0.3996, thereby indicating a distinct and substantial clustering structure. Additionally, the mean silhouette value for clustering was determined to be 0.755, implying that this particular clustering method is both rational and dependable. Based on different cluster labels, a total of 11 clusters were obtained: #0 stimulation, #1 mild cognitive impairment, #2 central mechanism, #3 state, #4 acupuncture analgesia, #5 research design, #6 neural regeneration, #7 alternative medicine, #8 ischemic stroke and #9 high-tech acupuncture as well as #10 sensation. The labels #0, #7, and #9 primarily pertain to intervention methods, indicating a diverse range of approaches for interventions and highlighting the need for further research in the field of alternative medicine on this specific topic. The primary focus of Labels #1 and #8 lies in brain-related diseases, while Labels #2, #3, #4, #6, and #10 are associated with research mechanisms. This suggests that scholars have begun to direct their attention toward the action mechanisms of acupuncture.

3.10 Analysis of keywords with the strongest citation bursts

Through burst analysis of the selected literature keywords, 25 burst terms were identified (Figure 6B). Early research primarily focused on foundational concepts such as “human brain,” “pain,” “humans,” “stimulation,” and “cortical activation,” with studies on the human brain being the earliest and most sustained. Mid-term research emphasized mechanism exploration, with keywords like “default mode,” “specificity,” “modulation,” and “neural regeneration” emerging. Later, the focus shifted to randomized controlled trials (RCT) to explore disease mechanisms and summarize the applications of fMRI in the field of acupuncture. During this phase, keywords such as “randomized controlled trial,” “meta-analysis,” “functional connectivity,” “ischemic stroke,” “women,” “systematic review,” “prophylaxis,” and “mild cognitive impairment” emerged, which remain significant today. These keywords are expected to become key areas of research focus and trends in the future.

4 Discussion 4.1 General information

This study employed Citespace and VOSviewer software to conduct a visual analysis of fMRI research in the field of acupuncture, encompassing a total of 557 articles. Through analyses of annual publications, collaboration networks, and Co-citation patterns, the latest advancements in fMRI research within the field of acupuncture were unveiled.

The distribution of annual publications shows that research on fMRI in acupuncture can be divided into three phases. The period from 2004 to 2010 is considered the initiation phase, with a relatively low number of publications and limited research in the field. During this phase, studies mainly focused on the mechanisms of acupuncture stimulation and pain. From 2010 to 2017, there was a steady upward trend in publications. Research during this phase mainly concentrated on neuroregeneration, neural specificity, and the default network. Since 2017, there has been a notable increase in the number of publications, with broader attention given to this area of research. Studies in this later phase have focused on randomized controlled trials, meta-analysis, systematic reviews, ischemic stroke in women, and the prevention of mild cognitive impairment. As the search strategy was conducted with a cutoff date of April 30, 2024, the publication volume for 2024 is lower. However, with 19 publications already published, it is expected that the publication volume for 2024 will at least match or even exceed that of 2023.

The Peoples R China has the highest number of publications, significantly outpacing the USA and other countries and regions, and holds the highest centrality, indicating a deep engagement with research in this field. This is likely related to China’s role as the birthplace of acupuncture. The USA and South Korea rank second and third in terms of publication volume, and notably, they also rank second and third in centrality, reflecting their broader research activities and the publication of influential articles. Germany and Taiwan (China) are ranked fourth and fifth in terms of publication volume. Compared to these four countries, Australia and Switzerland have published fewer studies on fMRI in acupuncture, but both countries show similar performance in centrality. Furthermore, there is limited exchange between countries, highlighting the need to strengthen international collaboration.

According to the collaborative network of institutions, it is evident that China holds four out of the top five positions in terms of publication volume. Similarly, when considering centrality, four out of the top five institutions are from China. These institutions primarily comprise research institutes or universities as their core teams. Kyung Hee University ranks fifth in terms of publication volume, yet it possesses a high centrality score of 0.18 > 0.1, indicating its significant research output and considerable influence in this field. Although Harvard University’s publication volume did not rank among the top five, its centrality score of 0.16 > 0.1 suggests that despite a relatively lower number of publications, their exceptional quality remains evident.

The collaborative network analysis results conducted by the author, Tian, J has emerged as the most prolific researcher, demonstrating an impressive publication count of 36. Closely following are Qin, W, Zeng, F, Bai, LJ, Kong, J and Liang FR. After analyzing the results of co-cited authors, it is found that Kong, J has the highest number of citations, followed by Hui, KKS, Napadow, V, Bai, LJ, and Dhond, RP.

Tian, J and Qin, W have mostly focused on acupoint specificity. They focused on changes in the brain area after needling acupuncture points. Zeng, F and Liang, FR have mostly focused on the central mechanisms of acupuncture for the treatment of diseases. Bai, LJ’s study focused on acupoint specificity, followed by more research on brain area changes after acupuncture treatment of disease. Much of Kong, J’s research has focused on the central mechanisms and brain region changes in the acupuncture treatment of disease. Hui, KKS research focused on brain networks and cortical plasticity. Napadow, V. and Dhond, RP’s research focused on brain networks and neuroplasticity.

The analysis of journal and co-cited journals reveals that Evidence-based Complementary and Alternative Medicine has the highest number of publications. The journal also ranks 4th in terms of citation count, indicating its significant contribution and dissemination of fMRI research in the field of acupuncture. Among the top 10 journals in terms of publication frequency, Neural Regeneration Research (IF = 5.9), Frontiers in Neuroscience (IF = 3.2), and Neural Plasticity (IF = 3.0) exhibit the highest impact factors. The majority of impact factors are below 5, and the JCR classification is mainly Q3, implying that fMRI research in acupuncture needs to be published in high-quality articles to enhance its influence. The journal with the highest number of citations is Neuroimage, which primarily falls into the Q1 and Q2 categories, indicating a superior level of quality and providing a solid foundation for fMRI research in acupuncture.

The analysis of journal and co-cited journals reveals that the most frequently published journal is Evidence-based Complementary and Alternative Medicine. Furthermore, this journal ranks fourth in terms of citation count, indicating that it has provided substantial support for research on fMRI in the field of acupuncture and has a certain level of dissemination impact. Among the top 10 journals by publication volume, the journals with the highest impact factors are Neural Regeneration Research (IF = 5.9), Frontiers in Neuroscience (IF = 3.2), and Neural Plasticity (IF = 3.0). Most of the journals have impact factors below 5, and the majority are categorized in the Q3 quartile according to JCR, suggesting that research on fMRI in acupuncture needs to be published in higher-quality journals to enhance its influence. The most frequently cited journal is Neuroimage, and the journals that are most frequently cited are primarily in the Q1 and Q2 quartiles, demonstrating their high quality and providing a solid foundation for research on fMRI in the field of acupuncture.

Co-citation analysis of references reveals that these cited studies are all clinical trials. The sample sizes of the included studies range from 10 to 36 participants, which, despite being relatively small, may be attributed to the high cost associated with fMRI. The most frequently cited reference is the 2008 article by Dhond et al. (2008), which demonstrated that acupuncture can enhance the spatial extent of resting-state brain networks, including brain regions related to pain modulation, memory processing, and emotional regulation. The second most cited reference is the article by Fang et al. (2009), which, through fMRI detection of real and sham acupuncture points, demonstrated that acupuncture leads to widespread deactivation of the limbic-paralimbic-neocortical system. The third and fourth most cited references are two articles by Bai LJ published in 2009. The third article Bai et al. (2009) demonstrated through fMRI findings that acupuncture can enhance the bipartite division of the anti-correlated resting-state networks and regulate the spontaneous activity of the intrinsic sensory-autonomic networks. The fourth article Bai et al. (2009) confirmed that brain activity patterns during rest are related to the type of stimulation. Significant stimulation-related activity was still present during the “rest” phase, particularly in the cerebellum and limbic brain regions. The fifth article Hui et al. (2005) used fMRI to explore how acupuncture at the Zusanli (ST36) point can modulate the neural activity of cortical and subcortical structures, including the cerebellum, diencephalon, and brainstem, at multiple levels. The highest centrality is held by the article by You et al. (2013), which used fMRI and MEG analysis to explore the specific biological mechanisms underlying acupuncture across spatial and temporal domains. This study found that modulation of the posterior cingulate cortex (PCC) within the theta (4–8 Hz), alpha (8–13 Hz), and gamma (30–48 Hz) frequency bands regulates the Default Mode Network (DMN) hub. The second most central reference is the article by Yeo et al. (2012), which used fMRI to show that acupuncture at GB34 (Yanglingquan) can increase neural responses in regions such as the substantia nigra, caudate nucleus, thalamus, and putamen, which are affected by Parkinson’s disease, confirming that acupuncture at GB34 may effectively improve symptoms of Parkinson’s disease. The third article is by Chen et al. (2015), which used fMRI to discover that acupuncture at GB34 can increase motor-cognitive connectivity while simultaneously reducing contralateral motor cortex compensation, thus promoting the rehabilitation of hemiplegia and spasticity. The articles by Chen et al. (2015), Li et al. (2010), and Wang et al. (2016) are tied for fourth. Chen et al. (2015) used fMRI to show that acupuncture can treat pain caused by knee osteoarthritis by enhancing functional connectivity between the right frontoparietal network (fFPN), the executive control network (ECN), and the descending pain modulation pathways. Li et al. (2010), conducted a multi-voxel pattern analysis (MVPA) based on fMRI data to explore the specificity of acupuncture at Guangming (GB37) and a nearby non-acupoint (NAP) on visual-related brain activation patterns. The study found that multiple brain regions could distinguish the central nervous response patterns caused by acupuncture at these two points, with the most significant areas being the occipital cortex, the limbic-cerebellar system, and subregions of the somatosensory cortex. Wang et al. (2016) conducted experiments that showed acupuncture can enhance resting-state functional connectivity between the left amygdala and the anterior cingulate cortex, as well as between the right amygdala and the left parahippocampal gyrus, thus helping to counter depression.

4.2 Research hotspots

By conducting keyword co-occurrence analysis and keyword clustering, the main research hotspots can be summarized as follows: different acupuncture techniques, acupoint specificity, and mechanisms of acupuncture in the treatment of diseases.

4.2.1 Different acupuncture techniques

Common intervention methods include electroacupuncture, acupuncture, and high-tech acupuncture, which encompasses techniques such as balance acupuncture, auricular acupuncture, and wrist-ankle acupuncture. Electroacupuncture is the most frequently used method. FMRI studies have shown that electroacupuncture induces a broader range of activation changes (Kong et al., 2007). Vitaly Napadow et al. (2005) suggested that all acupuncture stimuli result in a signal reduction in the amygdala, anterior hippocampus, the cortices of the subgenual and retrosplenial, and ventromedial prefrontal cortex, while signals increase in the anterior insula. However, compared to manual acupuncture, electroacupuncture at low frequencies induces a larger area of fMRI signal increase, with significant signals only observed in the anterior mid-cingulate cortex. Acupuncture is a widely used intervention. Wang et al. (2020) utilized fMRI to study the efficacy of multi-point acupuncture for treating primary insomnia and its effects on brain activity. High-tech acupuncture mainly refers to laser acupuncture, which is considered a safe and reliable option (Beissner et al., 2011). Existing studies have used fMRI to observe the impact of laser acupuncture on brain activity when stimulating acupuncture points (Lv et al., 2016; Chang et al., 2023).

4.2.2 Acupoint specificity

Stimulation of acupoints activates the brain Acupoints reflect the specificity of acupoints. Jung et al. (2015) found that both real acupuncture and fake acupuncture could activate the insula, anterior cingulate cortex, secondary somatic sensory cortex parietal cortex, and inactivate the medial prefrontal cortex, posterior cingulate cortex, inferior parietal cortex, and parahippocampal brain (Cao et al., 2019). Compared with shame acupuncture, real acupuncture showed greater brain activation in the posterior insula, posterior gill cover, and the tail of the anterior cingulate cortex. The study conducted by Ma et al. (2020) demonstrated the inhibitory effect of acupuncture on the marginal zone in rats with dysmenorrhea. Furthermore, acupuncture can modulate thalamocortical networks by enhancing connectivity between the mediodorsal thalamus (MD) and angular gyrus/posterior cingulate cortex/parahippocampal gyrus (PHG), as well as between motor thalamus subregions (Mthal) and medial prefrontal cortex (mPFC)/PCC/lateral temporal lobe cortical regions, while simultaneously reducing connectivity between MD and precuneus (PCU)/ middle cingulate cortex (MCC) (Kong et al., 2023). The findings of a review article suggest that acupuncture exerts an influence not only on the primary motor cortex and premotor cortex, but also leads to structural changes in language-related brain regions, including the inferior frontal gyrus, temporal lobe, occipital lobe, and parietal lobe (Zhang et al., 2021).

4.2.3 Mechanisms of acupuncture in the treatment of diseases

Functional connectivity can be used to study the mechanisms of acupuncture-induced analgesia. Pain is a widespread and complex symptom, and many patients heavily rely on analgesics. However, analgesics often present issues such as safety concerns and the potential for addiction (Qiao et al., 2020). In contrast, acupuncture is well-tolerated and generates fewer adverse effects. Acupuncture can alleviate pain by stimulating specific acupuncture points on the body (Chen et al., 2020; Robinson et al., 2022), which has made it increasingly popular. Neuroimaging techniques are essential for exploring the central mechanisms of acupuncture analgesia, with fMRI being able to reveal multifunctional brain networks associated with pain perception (Damascelli et al., 2022). Additionally, functional connectivity analysis has been used to explore the relationships between brain regions, helping to better understand analgesic mechanisms (Lee et al., 2019). In a study by Xiang et al. (2021), it was noted that ankle acupuncture alleviated chronic low back pain symptoms by modulating the cerebellum and right insula. Yang et al. (2022) used fMRI to explore the mechanisms of moxibustion in alleviating pain in patients with primary dysmenorrhea and found that moxibustion significantly increased the resting-state functional connectivity (rs-FC) of the left inferior frontal gyrus (IFG), bilateral anterior cingulate cortex (ACC)/middle cingulate cortex (MCC), left posterior cingulate cortex (PCC)/precuneus (PCU), and left parahippocampal gyrus (PHG). Wang et al. (2021) used fMRI to investigate the immediate analgesic effects of acupuncture on patients with primary dysmenorrhea, discovering that acupuncture significantly reduced the resting-state functional connectivity between the rostral anterior cingulate cortex (rACC) and left precentral gyrus. Wei et al. (2022) used fMRI to analyze changes in functional connectivity in patients with migraine without aura. The results showed that electroacupuncture at the rate of shuaigu (GB80) could modulate the functional connectivity between the right insular subregion and the parietal lobe, specifically the right dorsal anterior insula and the right postcentral gyrus, as well as between the right posterior insula and the left precuneus, thereby improving symptoms in patients with migraine without aura.

Neural regeneration is an important central mechanism for studying brain-related diseases. The regeneration of the nervous system necessitates the repair or replacement of impaired nerve cells resulting from injury or disease (Steward et al., 2013). Ischemic stroke is a sudden neurological deficit resulting from the cessation of blood flow to a specific region of the brain (Haupt et al., 2023). While acupuncture can stimulate neurogenesis, activate axonal regeneration and sprouting, improve synaptic structure and function, and enhance neural plasticity (Qin et al., 2022), especially when using appropriate electroacupuncture parameters for treatment, it can play a role in neuroprotection and promoting nerve regeneration (Chang et al., 2018). The loss of neurons and synapses is a primary pathological feature of cognitive impairment. Stimulating acupoints can restore brain functional activity and connectivity, promote neuroregeneration, and regulate synaptic plasticity to facilitate the formation of normal neural circuits (Zhang et al., 2023). FMRI can assess spontaneous neuronal activity through measures such as ALFF (Amplitude of Low-Frequency Fluctuations) and ReHo (Regional Homogeneity) (Zang et al., 2023). Changes in ReHo values can serve as one of the neurobiological markers reflecting disease (Zhang et al., 2018). Research by Li et al. (2022) demonstrated that electroacupuncture at Zusanli (ST36) and Quchi (LI11) significantly increased ReHo values in the ipsilateral posterior hippocampus, contralateral anterior hippocampus, contralateral subhippocampus, contralateral dorsolateral thalamus, ipsilateral somatosensory cortex, and ipsilateral substantia gelatinosa area, thereby promoting the activation of neurons in brain regions associated with cognition and behavior, reducing neuronal death. According to a comprehensive summary by Yin et al. (2023), acupuncture for cognitive disorders, through stimulation of 10 key brain areas such as the cingulate cortex, middle frontal gyrus, and hippocampus, can effectively promote neuroplasticity and facilitate repair. Zhang et al. (2024) utilized fMRI to investigate the underlying mechanism of acupuncture treatment for mild cognitive impairment (MCI). The findings revealed that following acupuncture intervention, MCI patients exhibited contrasting alterations in ReHo values within the hippocampus/parahippocampal and insular regions compared to baseline, while a notable increase in ReHo values was observed within the middle frontal gyrus region. These results suggest that acupuncture has the potential to modulate brain activity, particularly within the default mode network and salience network, thereby facilitating cognitive enhancement in individuals with MCI.

4.3 Research trend analysis

The emergence of keywords to a certain extent reflects the prevailing research trends. The graph reveals that keywords such as randomized controlled trial, meta-analysis, functional connectivity, ischemic stroke, women, systematic review, prophylaxis, mild cognitive impairment, and others have consistently maintained their popularity in current research directions.

4.3.1 Randomized controlled trials and functional connectivity

From the perspective of evidence-based medicine, well-designed randomized controlled trials (RCT) provide the essential information and guidance required for clinical practice decisions and are considered the gold standard, sitting at the top of the evidence pyramid. RCT is the most rigorous and reliable research method to determine whether there is a causal relationship between interventions and outcomes (Li et al., 2018; Bhide et al., 2018). Although fMRI has made significant progress in studying the safety and efficacy of acupuncture for treating diseases, the central mechanisms remain unclear and require further investigation. RCT is the best approach for this purpose. Functional connectivity can be used to explore the connectivity and interactions between brain regions within brain networks during specific stimuli (Xie et al., 2014; Liu et al., 2015). Therefore, using RCTs to explore brain functional connectivity can help reveal the underlying central mechanisms. Several RCT has shown (Liu et al., 2022; Li et al., 2022) that acupuncture treatment can enhance functional connectivity between specific regions when treating conditions such as anxiety and migraines, and that different doses of laser acupuncture lead to changes in functional connectivity between the regions of interest (ROI) and other brain areas (Chang et al., 2023). Additionally, an ongoing RCT is exploring the effectiveness and safety of acupuncture in treating post-stroke depression, particularly focusing on its impact on the functional connectivity of the cognitive control network (CCN) (Luo et al., 2022).

4.3.2 Meta-analysis and systematic review

Meta-analysis is a method that synthesizes, combines, and analyzes data from multiple studies to estimate a single effect and answer research questions (Arya et al., 2020), while systematic reviews are considered the gold standard for evidence synthesis (Paul and Leibovici, 2014). The literature included in this article related to meta-analysis and systematic reviews predominantly focuses on using fMRI to systematically summarize the mechanisms of acupuncture in treating brain disorders. Research by Zang et al. (2023) showed that fMRI revealed significant effects of acupuncture on the default mode network in patients with primary insomnia, particularly in the frontal and precuneus regions. A study by Zhang et al. (2021) demonstrated that after acupuncture treatment for depression, fMRI showed an increase in the N-acetylaspartate/creatine (NAA/Cr) ratio, increased ALFF in the right precuneus, decreased ALFF in the inferior frontal gyrus (IFG), and enhanced functional connectivity in the anterior cingulate cortex (ACC).

4.3.3 Ischemic stroke and women

Ischemic stroke is one of the most common and severe manifestations of cerebrovascular disease (Feske, 2021). Clinically, it presents as focal neurological deficits such as hemiplegia, aphasia, dysphagia, visual impairment, and cognitive impairments (Li et al., 2022). The incidence of stroke ranks second among the leading causes of death worldwide, accounting for 11.6% of all mortalities, and it also stands as the third most prevalent cause of disability globally, affecting 5.7% of the population (GBD 2019 Stroke Collaborators, 2021). Compared with males, females with ischemic stroke exhibit accelerated physiological aging, especially among younger populations where the incidence of ischemic stroke is higher in females than in males (Gallego-Fabrega et al., 2022; Ekker and de Leeuw, 2020). Additionally, females have a higher lifetime risk of developing ischemic stroke compared to males (Bruce et al., 2020). Acupuncture can promote neuroplasticity by modulating the functional reorganization of the entire brain after ischemia and by altering neural structure and function (Qin et al., 2022; Chavez et al., 2017) and has been used in the treatment of stroke and post-stroke rehabilitation (Lu et al., 2016). The main analysis methods utilized in fMRI