To facilitate clinical application, we further refined our results and proposed a neuroimaging-based protocol for each disorder by identifying the seven or eight surface regions based on the meta-analysis results. The 10–20 EEG system and the international standard acupoints were used to help locate the targets. We summarized the brain functions of each identified brain region associated with a corresponding neurological disorder to help the readers understand the specific brain functions of identified areas (Supplementary Tables 1c,2c,3c,4c,5c,6c,7c). Detailed descriptions of identified targets for each disorder based on 10–20 EEG system coordinates and acupuncture points can be found in Figs. 2,3,4,5,6,7,8 and Tables 1,2,3,4,5,6,7, respectively.

We investigated overlapping brain areas across different neurological disorders (see Table 8 and Fig. 9 for details). We found that: (a) Alzheimer’s disease, aphasia, chronic pain, dementia, dyslexia, and Parkinson’s disease show an overlap on the right supplementary motor area (SMA)/medial frontal gyrus (MedFG) (Fig. 9a); (b) the left lateral prefrontal cortex (LPFC) and inferior frontal gyrus (IFG)/orbital inferior frontal gyrus (OrbIFG) are involved in dementia, dyslexia, MCI and Parkinson’s disease (Fig. 9b); (c) the right middle frontal gyrus (MFG)/inferior frontal gyrus (IFG) are involved in Alzheimer’s disease, aphasia, and MCI (Fig. 9c); (d) Alzheimer’s disease and Parkinson’s disease display an overlap on the right superior temporal gyrus (STG)/Rolandic operculum (RO) (Fig. 9d); (e) the right medial prefrontal cortex (MPFC) are associated with aphasia and dementia (Fig. 9e); and (f) the right superior parietal lobule (SPL)/angular gyrus (AG) are involved in dyslexia and MCI (Fig. 9f).

Although each neurological disorder is associated with specific characteristic symptoms and underlying mechanisms, the boundaries between certain neurological disorders may be complex and contentious. For example, mild cognitive impairment, Alzheimer’s disease, dementia, and Parkinson’s disease present overlapping/progressing clinical symptoms, particularly in the aspects of memory and cognitive decline [42, 43]. As different brain regions or circuits are involved in different functions, symptoms may be regarded as the effects of dysfunction in certain areas or circuits. Evidence has shown that brain regions and circuit implicated in different neurological disorders overlap and clinical symptoms may be largely shared among disorders [44, 45]. Thus, identifying overlapping regions associated with different neurological disorders may not only be useful in illuminating the common etiology of different neurological disorders, but also help to identify common treatment targets across these disorders.

We found that the SMA is conspicuously involved in six disorders, including Alzheimer’s disease, aphasia, chronic pain, dementia, dyslexia, and Parkinson’s disease. Although the primary function of SMA is to control physical movement, a number of studies have found that the SMA is also a preferential site of several different neurological disorders. For instance, a recent study found that SMA damage affects working memory, thus, working memory impairment may be part of SMA syndrome [46], which commonly presents as a transient of speech and motor function disturbance [47]. Vergani and colleagues have suggested that the SMA plays a critical role in the control of motor aspects of speech production based on its connectivity with Broca’s area [48], which is consistent with the speech impairment of Alzheimer’s disease, dementia, aphasia, dyslexia, and Parkinson’s disease.

Although the SMA is believed to primarily be a motor area, it is also involved in pain processing [49]. In a previous study, Misra and Coombes executed an fMRI study while subjects performed a motor control task, experienced a pain-eliciting stimulus on their hand, and performed the motor control task while also experiencing the pain-eliciting stimulus [50]. They found that when separate trials of motor control and pain processing were performed, overlapping functional activity was detected in the SMA. Also, as motor control and pain processing occurred simultaneously, SMA activity increased.

The prefrontal cortex is a multi-functional brain area involved in working memory and social and emotional function regulation (including behavioral control, decision making, etc.). We found that the left lateral prefrontal cortex (LPFC) is a notable region that contributes to four disorders: dementia, dyslexia, mild cognitive impairment, and Parkinson’s disease. Literature suggests that the LPFC has long been implicated in higher cognitive functions, such as attention switching and working memory formation [51, 52]. To explore the treatment effects of tDCS on patients with Parkinson’s disease, researchers applied anodal stimulation on the left LPFC and found a significant improvement of patients’ working memory [53], whereas anodal stimulation on both the left and right LPFC positively impacted the executive function [54]. Moreover, anodal tDCS targeted on Broca’s area (a brain region of the dominant hemisphere that adjacent to the LPFC with functions related to speech production) can produce positive effects on verbal fluency in patients with mild cognitive impairment [55, 56], this effect may also benefit patients with other disorders presenting speech impairments.

We also found an aphasia and dementia overlap in the medial prefrontal cortex (MPFC). A large body of evidence indicates that the MPFC is essential for theory of mind [57], response inhibition [58], reversal-learning [59], and emotional processing [60]. MPFC dysfunction is an early marker of behavioral variant frontotemporal dementia (bvFTD), the most commonly occurring subtype of frontotemporal dementia as characterized by progressive changes in personality and impaired social interaction [61]. In a previous study, Bertoux et al. [62] investigated the sensitivity and specificity of common MPFC specific tests in patient cohorts (bvFTD and Alzheimer’s disease) and age-matched healthy individuals. They found that the Mini-SEA test, which was employed to evaluate theory of mind and emotional processing, emerged as the most sensitive and specific of the MPFC tests employed, and may be used to discriminate bvFTD and Alzheimer’s disease.

In addition, our results showed that Alzheimer’s disease, aphasia, and mild cognitive disorder overlap in the middle/inferior frontal gyrus (MFG/IFG). These disorders present varying levels of language defects. Evidence for the MFG/IFG as a structure essential to speech and language function has been verified repeatedly through accumulating studies across a variety of neuroimaging approaches including functional magnetic resonance imaging [63], magnetoencephalography [64], positron emission tomography [65], and single-photon emission computed tomography [66]. These studies indicate that the MFG/IFG is involved in multiple language-specific tasks including phonologic, semantic, and sentence/discourse-level processing, as well as detection of the emotional content of speech [67]. Thus, the MFG/IFG possess the potential to be a scalp stimulation target in modulating disorders which exhibit language/speech-related disabilities.

We found the superior temporal gyrus (STG) and the Rolandic Operculum (RO) may be considered potential brain region targets in treating Alzheimer’s disease and Parkinson’s disease (two of the most common neurodegenerative diseases). A previous neuroimaging study found that Parkinson’s disease involves greater grey matter loss in frontal areas and the temporal lobe in Alzheimer’s disease relative to Parkinson’s disease dementia (PDD) [68]. These findings are consistent with another MRI study, in which researchers applied voxel-based morphometry (VBM) analysis in a group of PD patients with and without depression and reported gray matter volume reduction in the orbitofrontal cortex and the STG of PD patients with depression symptoms [69]. To investigate the clarification of regional morphologic changes in the brain associated with normal aging and AD, Ohnishi and colleagues perfromed an MRI study on 26 AD patients and 92 healthy individuals [70]. Negative correlations between age and regional gray matter volume of STG have been detected in healthy individuals (i.e., significant reduction of STG volume with age). Swallowing impairment is a growing concern in both AD [71] and PD patients [72], particularly in the disease’ later stages [73]. Numerous studies have suggested that the operculum, a critical cortical region in involved in normal swallowing, is affected by these diseases [74, 75]. For instance, Humbert et al. [73] revealed that AD patients had a significantly lower Blood-Oxygen-Level-Dependent (BOLD) response in many cortical areas that are traditionally involved in normal swallowing, especially in the Rolandic and frontal operculum.

Lastly, we found that dyslexia and mild cognitive impairment (MCI) display an overlap on the superior parietal lobe (SPL) and angular gyrus (AG). Converging evidence, from a number of lines of investigation, indicates that dyslexia may represent a disorder within the language system; such reading difficulties are also presented as prominent symptom in patients with MCI [76]. The French neurologist Dejerine highlighted the importance of the angular gyrus and the parietal lobe in reading performance as early as 1891 [77]. In a previous study, researchers found that dyslexic subjects showed a significant brain volume reduction in the right SPL compared to healthy controls [78]. Pugh et al. [79] also found that the AG is involved in reading dysfunction in dyslexic readers. Additional in vivo imaging studies have also linked dyslexia to abnormalities in the structures associated with the parietal and angular gyrus [80].

The stimulation targets we selected from the current study are partly consistent with the published prescriptions from scalp acupuncture (or traditional acupuncture), and neuromodulation studies [23, 81, 82, 83, 84, 85].

Take Parkinson’s disease for example, literature suggests that the MS 4 (Epangxian III, line 3 lateral to forehead,), MS 6 (Dingnie Qianxiexian, anterior oblique line of vertex-temporal), MD8 (Dingpangxian I, line 1 lateral to vertex, n), MS 9 (Dingpangxian II, line 2 lateral to vertex,), and MS 14 (Zhenxia Pangxian, lower-lateral line of occiput) are widely recognized for alleviating symptoms of Parkinson’s disease when using scalp acupuncture [6]. Overlapping regions exist in comparison of the neuroimaging-based targets and literature-documented locations, involving the SMA, PreCG, PoCG, SMG, and IPL. We also included the frontal gyrus (SFG/MFG/MedFG/IFG) into our proposed protocol. In parallel, as suggested by the literature, the occipital gyrus may be considered as a potential target in treating Parkinson’s disease [6], but our protocol did not include the occipital area.

In terms of neuromodulation interventions, most of the previous studies applied different neuromodulation techniques at primary motor cortex (M1) to improve the motor disability of PD. A recent literature review summarized the therapeutic application of repetitive transcranial magnetic stimulation (rTMS) in patients with Parkinson’s disease and found that stimulation targets of the twenty included research studies mainly targeted on the M1, SMA, dorsolateral prefrontal cortex (dlPFC), and MPFC [86]. In addition, tDCS on the M1 and dlPFC have been shown in several studies to ameliorate gait function in PD patients [87, 88]. Furthermore, patients with PD showed positive improvement of cortico-muscular coupling and motor performance after receiving 20 Hz tACS on the M1 [89].

In comparison with above studies, our targets also include additional brain regions, such as PoCG, Rolandic operculum, SMG, and IPL, which may expand the selection of potential targets in neuromodulation techniques for the treatment of Parkinson’s disease.

In the development of brain stimulation, how and where to stimulate remains a challenge for clinicians and researchers. In this study, we provide a set of stimulation location protocols for seven common neurological disorders based on the development of brain research using the Neurosynth. We believe that the brain regions included in the neuroimaging-based protocol reflect an enhanced understanding of the neural network involved in neurological disorders and thus, stimulating these brain regions using different techniques may hold the potential to regulate the pathophysiology associated these disorders and further relieve associated symptoms. Nevertheless, this assumption/hypothesis needs to be evaluated by additional clinical studies.

We provided multiple targets for each neurological disorder. Some of these targets may not have been used in previous studies, which provides more options for brain stimulation in neurological disorders. Readers may choose the targets based on symptoms associated with each individual patient and the function of different targets. To facilitate the reader to choose targets, we tried to summarize the functions of each identified brain regions; due to the complexity of the brain, we are almost certain these listed functions are incomplete and under certain circumstance, even may be misleading. The readers should make their judgement with cautions.

There are several limitations to our study. Since the protocol is based on the brain imaging meta-analysis, further clinical studies are needed to validate our findings. Additional functional (resting-state functional connectivity) [20, 21], and anatomical analyses (diffusion tensor imaging) [23, 90] may further enhance the proposed protocols, particularly in the target for patients at the individual level. The aim of this study was to explore potential targets for neurological disorders; the application and optimization of different scalp stimulation techniques to modulate these brain regions is beyond the scope of this manuscript. In order to facilitate clinical application, we simplified the protocol by including only seven or eight peak targets. Other brain regions may also play a critical role and should be considered in clinical practice. Finally, the exact location may alter if more studies are included in the analysis. Since scalp acupuncture may influence a relatively large area, the change should not influence the clinical effect significantly. Nevertheless, the protocol should be updated with the enhancement of our understanding of the pathology of these disorders and the brain function.