The keyword co-occurrence analysis reflects the primary research focuses within this field. Using citation frequency as the primary metric for observation, the ten most frequently cited keywords, excluding subject terms, are: “deep brain stimulation,” “subthalamic nucleus,” “movement disorders,” “surgery,” “thalamic stimulation,” “nucleus,” “thalamotomy,” “electrical stimulation,” “stimulation,” and “disease” (Figure 6; Table 3).

The clustering diagram of co-occurring keywords visually illustrates the interconnections and thematic groupings of these research areas. The Q and S values in the clustering diagram serve as measures of result reliability; a Q value greater than 0.3 and an S value greater than 0.7 indicate a reliable clustering result. For this analysis, the Q value for keyword clustering is 0.487, and the S value is 0.7773, demonstrating the reliability of these clustering outcomes. This analysis successfully identified 14 distinct cluster labels: #0 thalamotomy, #1 dystonia, #2 transcranial magnetic stimulation, #3 local field potentials, #4 deep brain stimulation, #5 diffusion tensor imaging, #6 harma line, #7 magnetic resonance imaging, #8 design, #9 larynx, #10 devices, #11 high-frequency stimulation, #12 direct current stimulation, and #13 cerebello-thalamocortical pathway (Figure 7).

The burst analysis of keywords clearly reflects the research hotspots in this field over specific time periods (Figure 8). This analysis can be divided into three phases: the first phase, from 2004 to 2012, identified hotspot keywords primarily focused on “proximal tremor, neurons, multiple sclerosis, follow up, intention tremor, globus pallidus internus, subthalamic nucleus stimulation, dystonia.” This phase primarily discussed the improvement of tremor symptoms due to surgical interventions. The second phase, from 2013 to 2018, recognized hotspot keywords that included “term follow up, disruption, prevalence, focused ultrasound, functional neurosurgery, focused ultrasound thalamotomy, gamma knife thalamotomy.” During this period, focused ultrasound thalamotomy and gamma knife thalamotomy began to gain prominence. The third phase, from 2018 to the present, comprises keywords such as “trial, DBS, network, caudal zona incerta, consensus statement, ventral intermediate nucleus, location, outcome, MR-guided focused ultrasound, Parkinson.” Both deep brain stimulation and magnetic resonance-guided focused ultrasound remain predominant research areas, while the selection of surgical location is attracting increasing attention from researchers.

DBS involves delivering electrical impulses to specific nuclei in the brain (e.g., VIM or PSA) through a drilled hole in the skull while the patient is under local anesthesia, aiming to regulate abnormal neural signals to suppress or alleviate tremors (24). Since its FDA approval in 1997, DBS has played a crucial role in tremor treatment. Unlike radiofrequency ablation or gamma knife radiosurgery (25), DBS does not destroy brain tissue, thereby minimizing the risk of permanent damage. The effects of DBS are immediate, with therapeutic outcomes observable shortly after surgery.

However, DBS is not a permanent solution; parameters must be continuously adjusted in response to disease progression and symptom changes (26). Consequently, automation has been introduced in this area. Researchers (27–29) have implemented real-time detection of neural signals to automatically adjust stimulation intensity and other parameters based on patient needs. An innovative fully implanted bidirectional neural interface can detect neural activity during stimulation and includes circuits for feedback control (29). However, this technology is not yet fully developed, and our understanding of various brain signals remains limited—this understanding is a prerequisite for automated adjustments (30). Additionally, algorithms for adjusting stimulation frequency still pose a challenge, necessitating further research for optimization (31).

Recently, MR-guided focused ultrasound thalamotomy has emerged as an important research focus in essential tremor. Among the ten most cited papers in this area, six pertain to MR-guided focused ultrasound thalamotomy. The study “A Randomized Trial of Focused Ultrasound Thalamotomy for Essential Tremor” by Elias et al. (16), which garnered the highest number of citations, details the enrollment of 76 patients with essential tremor (ET) across eight international research centers. These patients were randomly assigned in a 3:1 ratio to undergo either MRgFUS or a sham procedure. In the MRgFUS group, the hand tremor score improved from a baseline of 18.1 to 9.6 at 3 months, whereas the sham group’s score changed marginally from 16.0 to 15.8. The improvements in the MRgFUS group were maintained at the 12-month mark.

MRgFUS thalamotomy is a minimally invasive procedure that can be performed while the patient is awake, thereby reducing surgical risks (32). The therapeutic effects of MRgFUS thalamotomy are reversible, allowing for repeat treatments or adjustments to the treatment area if necessary. This technique presents broader indications, offering an alternative treatment option for patients unable to accept the risks associated with DBS, those with concerns about implanted devices, or individuals who are unsuitable candidates for DBS (12, 33).

However, MRgFUS thalamotomy also has its limitations. Bone density and structure may interfere with ultrasound transmission, and patients with intracranial metal are not suitable for MRI scans. Furthermore, caution is warranted in patients with cognitive or psychiatric disorders (12, 33). On the other hand, MRgFUS thalamotomy improved hand tremor scores by 47% at 3 months, with this improvement lasting for 12 months (16). Bilateral VIM DBS can reduce overall tremors by 66–78%, with improvement rates of 60.3–75% for upper limb action tremors 5 years post-surgery. Consequently, the efficacy of MRgFUS thalamotomy is somewhat inferior to that of DBS, and there is a lack of long-term efficacy follow-up studies on MRgFUS thalamotomy. Large-sample, multicenter, and long-term evaluation studies are needed.

DBS targeting the ventral intermediate nucleus (VIM) is a well-established, effective, and safe treatment method (34). However, recent studies have identified the posterior subthalamic area (PSA) (35)and caudal zona incerta (cZI) (36, 37) as potentially more advantageous stereotactic targets. In the highly-cited literature, two articles discussed the clinical benefits of different targets. Barbe et al. (11) found PSA-DBS to be more effective than VIM-DBS, producing optimal stimulation effects at lower amplitudes and providing at least equivalent tremor control, with no differences in the quality or quantity of side effects between the two targets. However, side effects associated with PSA-VIM stimulation were more prominent regarding emotional and behavioral disturbances, likely due to PSA’s involvement in emotional regulation (38). Another relevant study conducted a double-blind crossover trial comparing VIM DBS with PSA DBS (39), noting a wiring trajectory that crossed both regions, with a dorsal contact point in VIM, a neutral contact point on the intercommissural line (ICL), and a ventral contact point in PSA. Follow-up after 12 months revealed no statistically significant difference in tremor suppression between VIM and PSA, although there was a slight trend toward better tremor suppression with PSA stimulation.

Some researchers propose that the efficacy of PSA and VIM DBS in essential tremor may depend on the distance to the dentate nucleus. Dembek et al. (35) conducted a randomized crossover trial involving 13 essential tremor patients implanted with contacts in both VIM and PSA. Three months post-implantation, they calculated the Euclidean distance from the electrode contacts in PSA and VIM to the center of the dentate nucleus using patient-specific MRI images aligned with a population-averaged DRTT model. The results demonstrated that the PSA contacts were closer to the dentate nucleus, resulting in superior tremor suppression on TRS scores.

cZI is another alternative target for ET DBS. The zona incerta is an area located beneath the thalamus that functions in the regulation of movement and pain transmission. Stimulation of cZI is believed to influence the entire motor network, including circuits from the cerebellum and basal ganglia, potentially improving tremor symptoms through modulation of movement-related neurotransmission (36). Eisinger et al. (40) conducted DBS on 47 essential tremor patients and found that the short-term benefits (after 6 months and 2 years) of VIM and cZI were similar. However, long-term follow-up (3–4 years) revealed a decreasing trend in tremor scores over time for VIM DBS, whereas cZI DBS exhibited an increasing trend; thus, cZI DBS appeared less effective in long-term tremor control compared to VIM DBS. Moreover, cZI DBS may lead to side effects related to cerebellar and motor control, such as unsteadiness or motor incoordination (41), underscoring the need for effective monitoring and management strategies to mitigate these adverse effects. Furthermore, the surgical procedure for cZI DBS is more challenging and complex due to its location beneath the thalamus and proximity to other critical neural structures, making precise target localization a significant challenge.

Additionally, the efficacy of MRgFUS may also depend on the location of the target. Arcadi et al. (12) found that the effectiveness and acute adverse reactions of MRgFUS largely rely on the precise location and size of the focused ultrasound thalamotomy.

Accurate targeting of the surgical site is critical for successful outcomes in both DBS and MRgFUS thalamotomy. There is a variety of localization techniques, including ventriculography, stereotactic frames, and MRI guidance. However, as research on targets becomes more in-depth, more complex targets (such as PSA and cZI) require more precise localization to determine their positions.

Target structures such as PSA and cZI are complex and challenging to localize, necessitating advancements in localization technology for the application of new DBS targets. Given current developmental and practical needs, specific breakthrough directions are warranted. Firstly, due to individual anatomical variations in brain structures, a universal stereotactic frame may not precisely accommodate each patient’s unique anatomy. Therefore, utilizing patient-specific MRI data to create personalized brain models can enhance localization accuracy. Secondly, postoperative brain shift is a significant concern; the brain tissue may shift following surgery, potentially disrupting the alignment between preoperative imaging and actual brain tissue (42). Thus, the development of imaging techniques capable of providing real-time feedback is essential. Lastly, artificial intelligence has become a burgeoning area of interest. Through advanced image processing techniques, pattern recognition, and machine learning algorithms, AI can accurately analyze and identify brain structures, optimize fiber tracking, compensate for real-time brain tissue displacement, and personalize surgical planning, significantly improving localization precision in deep brain stimulation surgeries.