Research hotspots and frontiers of neuromodulation technology in the last decade: a visualization analysis based on the Web of Science database

Neuromodulation is defined as the alteration of nerve activity through targeted delivery of stimuli (electrical, chemical, or other) to specific neurological sites. It aims to improve neural function and quality of life, spanning invasive approaches (e.g., deep brain stimulation, DBS) and non-invasive techniques (e.g., transcranial direct current stimulation, tDCS; transcranial magnetic stimulation, TMS; transcranial focused ultrasound stimulation, tFUS). Since the 1990s, neuromodulation has advanced rapidly and is used clinically for Alzheimer’s disease, stroke, and Parkinson’s disease, among others, offering advantages in precision and intelligence compared to traditional rehabilitation. To inform and guide future work, this study uses VOSviewer and CiteSpace to visualize and analyze neuromodulation research indexed in Web of Science from 2014 to 2024, examining co-authorship, citation, co-occurrence patterns, co-citation networks, and thematic evolution, with the goal of identifying international hotspots and development trends.

Co-cited literature analysis indicates that highly cited works were largely published before 2020, with ultrasound neuromodulation dominating the top ranks. Legon et al. (2014) demonstrated that tFUS can modulate human sensory-evoked brain activity and cortical function. Folloni et al. (2019) showed manipulation of subcortical and deep cortical activities in primates using tFUS, and Blackmore et al. (2019) provided a comprehensive review of ultrasound neuromodulation. Additional influential studies include Lee et al. (2015) on image-guided tFUS of the human somatosensory cortex, Legon et al. (2018a, 2018b) on thalamic and motor cortex tFUS neuromodulation, Verhagen et al. (2019) on offline tFUS effects in primates, and Deffieux et al. (2013) on focused ultrasound modulating visuomotor behavior. Collectively, these works highlight the emergence and growing recognition of focused ultrasound as a promising neuromodulation modality.

Data source: Web of Science Core Collection. Timeframe: January 1, 2014 to June 18, 2024. Search query: TS = (non-invasive neuromodulation) OR TS = (invasive neuromodulation) OR TS = (neuromodulation technology). Inclusion criteria: (1) Literature relevant to neuromodulation technology (non-invasive and invasive); (2) Clinical trials, reviews, meta-analyses, observational studies, systematic evaluations, and animal experiments; (3) English-language articles. Exclusion criteria: (1) Studies not aligned with the research theme; (2) Conference abstracts, reports, news items, or documents lacking sufficient information or duplicates. Screening: Two researchers independently screened titles, abstracts, and full texts, with disagreements resolved by a third researcher; included records were exported as plain text files. Tools and parameters: Data were imported into CiteSpace 6.2.R4 and VOSviewer 1.6.20 with a time span January 2014–December 2024. CiteSpace settings: time slice = 1 year; link strength = Cosine; node selection = Top N (N adjusted by data); used for co-cited references, keyword clusters, and burst detection. VOSviewer settings: analysis type = co-authorship and co-occurrence; full counting method; visualizations included network, overlay, and density maps; used for country/region, institutions, authors, and keywords analyses.

Publications: 1,424 initially retrieved; 1,348 included after screening. Annual trends: steady increase from 2014 to 2023, with the most significant growth between 2020 and 2021; by 2023, volume was ~4x that of 2014; average 2014–2023 = 120.8 publications/year. Countries/regions: 66 countries involved; top outputs—United States 631, China 206, United Kingdom 127, Italy 107, Germany 105, Canada 93, Spain 79, Belgium 63, Netherlands 54, South Korea 54; dense collaborations among US, China, and European countries. Institutions (top 10 by publications): Harvard Medical School 48; University of Toronto 46; Stanford University 35; UCLA 34; University of Minnesota 31; Mayo Clinic 27; Capital Medical University 23; University of Oxford 23; Columbia University 22; UCSF 21. Authors (top 10): Marom Bikson 15; V. Reggie Edgerton 12; Andres M. Lozano 10; Felipe Fregni 9; Scott F. Lempka 9; Eric Liebler 9; Long Meng 9; Parag Gad 8; Peter J. Goadsby 8; Zhengrong Lin 8. Co-cited literature (top): Legon et al., 2014 (30); Folloni et al., 2019 (26); Blackmore et al., 2019 (21); Lee et al., 2015 (19); Legon et al., 2018b (13); Verhagen et al., 2019 (12); Wagner et al., 2018 (12); Legon et al., 2018a (12); Deffieux et al., 2013 (11); Guo et al., 2018 (8). Keywords co-occurrence (top 10): deep brain-stimulation 170; transcranial magnetic stimulation 149; electrical-stimulation 124; tDCS 117; neurostimulation 90; cortex 78; brain-stimulation 76; therapy 70; excitability 67; activation 66. Clustering: seven clusters—#0 tDCS (size 55; silhouette 0.785), #1 chronic pain/SCS (size 40; silhouette 0.759), #2 sacral neuromodulation (size 36; silhouette 0.81), #3 focused ultrasound (size 35; silhouette 0.616), #4 deep brain stimulation (size 32; silhouette 0.745), #5 occipital nerve stimulation (size 13; silhouette 0.831), #6 TMS/tDCS/NIBS (size 12; silhouette 0.745); overall Q = 0.437 and S = 0.7526 indicating significant, high-quality clustering. Bursting keywords: earliest—cluster headache, peripheral neuromodulation, major depressive disorder; recent (past 3 years)—model, movement, plasticity; top burst strengths—disease (7.95), connectivity (6.03), nucleus (5.5), TMS (5.29), cluster headache (5.05). The analysis highlights central neuromodulation techniques (DBS, TMS, tDCS, tFUS) and pain management (including SCS) as core foci, with emerging interest in recovery, deep nuclei, modeling, and neuroplasticity.

The study maps a decade of neuromodulation research, demonstrating growing global interest and output, particularly from the US, China, and Europe, and from leading institutions such as Harvard Medical School and the University of Toronto. Findings identify DBS, TMS, tDCS, and tFUS as dominant modalities, with chronic pain management (e.g., spinal cord stimulation) a sustained hotspot. Advances include adaptive/closed-loop systems (DBS and SCS), neuronavigation and EEG-guided optimization for TMS, high-definition tDCS, and remote DBS programming, all pushing towards precision, personalization, and safety. Co-citation patterns emphasize focused ultrasound’s rise, with human and primate studies showing modulation of cortical and subcortical targets and functional connectivity changes, suggesting potent non-invasive access to deep structures. Burst analyses indicate future emphasis on motor recovery and neuroplasticity, deep brain targets (e.g., nucleus accumbens, striatum), modeling, and closed-loop, biomarker-guided strategies. Integration with imaging, electrophysiology, rehabilitation robotics, and brain–computer interfaces is poised to further enhance clinical translation and functional outcomes.

Neuromodulation techniques have advanced rapidly over the last decade. Key central techniques include DBS, TMS, tDCS, and tFUS, with peripheral approaches such as sacral neuromodulation and occipital nerve stimulation. The integration of imaging and physiological signals to improve safety, localization, and closed-loop personalization is a major trend. Future research will likely focus on exploring deep brain stimulation targets and on restoring motor function via neuroplasticity-oriented strategies to address movement disorders.

The search strategy and scope may have excluded relevant literature, including some technologies such as optogenetics. The analysis was limited to the Web of Science Core Collection and did not include other databases like PubMed. Variability and non-standardization in visualization parameters (e.g., time slicing, thresholds) may introduce bias.