Clinical Utility of Magnetoencephalography in Epilepsy Evaluation: A Qualitative Systematic Review

Abstract

Magnetoencephalography (MEG) is a non-invasive neurophysiological technique offering high spatial resolution for localizing epileptogenic zones in epilepsy, especially when traditional electroencephalography or magnetic resonance imaging (MRI) is inconclusive. A systematic evaluation of MEG’s diagnostic and prognostic utility within combination strategies is crucial, particularly in countries like South Korea with limited MEG access. We conducted a qualitative systematic review of nine studies (n=354 focal epilepsy patients) to evaluate MEG’s clinical performance in presurgical workup. Databases (MEDLINE, EMBASE, Cochrane, KoreaMed, KMbase, RISS) were searched. Data extraction focused on localization accuracy and surgical outcomes (Engel class I); risk of bias was assessed using quality assessment of diagnostic accuracy studies-2. MEG alone achieved up to the mid-70% range; however, integration with other modalities (e.g., with positron emission tomography/high-density electroencephalography) significantly improved both localization and surgical outcomes. Pediatric focal cortical dysplasia patients showed Engel class I outcomes of 67–87%. Most studies had low-to-moderate bias. Only one MEG system is operational in South Korea (introduced 2023), limiting accessibility. Canadian economic evaluations, despite higher initial costs, suggest MEG is long-term cost-effective, improving quality-adjusted life years. MEG offers complementary diagnostic value in epilepsy evaluation and surgical planning, enhancing localization and outcome prediction, especially for pediatric and MRI-negative patients. Considering this clinical utility, national support for MEG equipment and its regional expansion in South Korea is crucial to ensure equitable access and optimal patient care.

Introduction

Epilepsy is a complex neurological disorder affecting over 50 million people worldwide with approximately 30% of patients exhibiting drug-resistant forms.1,2 For these individuals, surgical intervention can be an effective treatment strategy, but its success hinges on accurate localization of the epileptogenic zone.3

From a patient’s perspective, accurate localization is paramount, as imprecise identification of the epileptogenic zone can lead to surgical failure or unnecessary neurological deficits.4 Conventional neuroimaging techniques such as magnetic resonance imaging (MRI) and electroencephalography (EEG) are commonly used; however, their sensitivity and specificity can be limited, particularly in MRI-negative cases or extratemporal epilepsy.5

Magnetoencephalography (MEG) is an advanced imaging technique that detects magnetic fields produced by neuronal activity. It provides a high temporal and spatial resolution and is particularly sensitive to tangential sources of cortical activity. This sensitivity allows MEG to identify epileptogenic zones with greater precision in certain cases compared to EEG, which primarily detects radial dipoles. However, in clinical practice, MEG is rarely used as a standalone modality for localizing epileptogenic zones in epilepsy surgery; instead, it is typically performed in conjunction with other non-invasive examinations to provide comprehensive insights.6

This multi-modality approach is crucial because each neuroimaging technique offers unique advantages and limitations, necessitating their combined use for optimal diagnostic yield. While the importance of multi-modality evaluation in presurgical planning is widely acknowledged, a comprehensive synthesis specifically examining the incremental value and synergistic effects of MEG when combined with other diagnostic tools remains limited.7 As a non-invasive tool, MEG offers a crucial advantage by potentially reducing the need for more invasive procedures like intracranial EEG (iEEG), thereby minimizing patient risks and discomfort, while still providing critical localization information.8

In South Korea, MEG was introduced into clinical practice in 2023 with support from the Ministry of Health and Welfare and is currently reimbursed under the national health insurance system with 80% coverage as a selective benefit.9–12 Despite this policy support, and with only one MEG system currently available nationwide,13,14 the clinical utilization of this advanced technology has not yet reached its full potential.

This recent policy change, coupled with the underutilization of the available equipment, has increased clinical access to MEG but also created an urgent need to systematically evaluate its real-world utility and evidence base in the Korean epilepsy care context.

Furthermore, the need for national support for epilepsy care, including advanced technologies like MEG, has been emphasized to ensure equitable access and optimal patient outcomes across the country.9

The present study aims to conduct a qualitative systematic review of the literature to evaluate the diagnostic and prognostic utility of MEG in epilepsy evaluation, focusing on its role in the presurgical workup. The primary outcome domains assessed were 1) the localization accuracy of MEG, both as a standalone modality and in combination with other modalities and 2) the surgical success rate (typically assessed by Engel class I outcomes) following MEG-informed or MEG-guided resection strategies.

Methods

This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. A comprehensive literature search was conducted across three international databases (Ovid-MEDLINE, Ovid-EMBASE, and Cochrane Central Register of Controlled Trials) and three Korean medical databases (KoreaMed, KMbase, and RISS) through August 26, 2024.

Search terms included “magnetoencephalography”, “MEG”, “epilepsy”, “seizure localization”, and “surgical outcome”. Studies were included if they reported the use of MEG in the presurgical evaluation of epilepsy, with a particular focus on its role within assessment strategies-where it was combined with other diagnostic techniques such as EEG, positron emission tomography (PET), or single-photon emission computed tomography (SPECT)-or where it provided complementary information. Studies examining MEG as a standalone modality were also included where relevant for comparative analysis.

Inclusion criteria were: 1) original peer-reviewed research articles, 2) studies involving human subjects, 3) application of MEG in epilepsy evaluation, and 4) studies investigating the combination or complementary use of different diagnostic modalities. Exclusion criteria included non-English or non-Korean articles, case reports with fewer than five patients, animal studies, and reviews.

Data extraction focused on MEG localization accuracy, surgical outcome (e.g., Engel classification), and concordance with other modalities. Two reviewers independently screened titles and abstracts, followed by full-text assessment for eligibility. Discrepancies were resolved by discussion and consensus.

Risk of bias was assessed using the quality assessment of diagnostic accuracy studies-2 (QUADAS-2) tool,15 which is designed to evaluate diagnostic accuracy studies. This tool considers four domains: patient selection, index test, reference standard, and flow/timing. Each study was independently rated by two reviewers.

Given the heterogeneity in study designs, patient populations, and reported outcomes, a narrative synthesis (qualitative systematic review) approach was adopted rather than a quantitative meta-analysis. This allowed for a detailed exploration of the diagnostic and clinical value of MEG in diverse clinical scenarios. To visualize and compare the clinical performance of MEG across patient subgroups, representative values for localization accuracy and Engel class I outcomes were extracted from each included study. Where multiple values were reported, a central or upper-bound estimate was used to reflect the general trend within that subgroup. These estimates were then used to construct a comparative bar graph summarizing MEG performance by age group, MRI status, and epilepsy type. This descriptive approach aimed to highlight clinically relevant patterns, while acknowledging variability across individual studies.

Results

Study characteristics

A total of nine studies met inclusion criteria, comprising both pediatric and adult patients with focal epilepsy (Appendix Table 1). Study designs included both retrospective and prospective analyses, and most studies compared MEG either as a standalone modality or in combination with other modalities such as high-density EEG (HD-EEG), PET, or iEEG. MEG was used in the evaluation of both MRI-positive and MRI-negative cases, as well as in patients with temporal and extratemporal lobe epilepsy (ETLE). The selection process is illustrated in Fig. 1, following PRISMA 2020 guidelines. A total of nine studies were included after screening 4,811 records and applying eligibility criteria.

Figure 1.

PRISMA flow diagram. DB, database; MEG, magnetoencephalography; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Risk of bias

Risk of bias across studies, assessed using the QUADAS-2 tool, is summarized in Fig. 2. Most studies demonstrated a low to moderate risk of bias. The most common concerns were related to patient selection, unclear reporting of index test interpretation, and variability in the timing between MEG evaluation and reference standard outcomes.

Figure 2.

Risk of bias summary (quality assessment of diagnostic accuracy studies-2).

Localization accuracy of MEG

Five studies reported quantitative data on MEG’s localization accuracy, both as a standalone modality and in combination with others (Table 1). While MEG alone demonstrated localization accuracy ranging from 57.3% to 75.3%, its clinical utility was significantly enhanced when integrated into assessments. When combined with other modalities, particularly HD-EEG or PET, localization accuracy improved markedly, reaching up to 93.2%. Sensitivity and specificity varied across studies. For instance, Chikara et al.16 demonstrated that combining HD-EEG with MEG yielded a Youden’s index of 70.63% and an overall diagnostic accuracy of 91.3% for surgical resection site prediction, superior to HD-EEG alone (41.27%) or MEG alone (33.33%).

Table 1.

MEG localization accuracy of studies

No.	Study	MEG alone accuracy	MEG+other accuracy	MEG alone		MEG+other		Notes
				Sn	Sp	Sn	Sp
1	Chikara et al. (2023)16	65.2	82.6	85.7	33.3	92.9	66.7	MEG+HD-EEG (SOZ ref.)
		69.6	91.3	85.7	44.4	92.9	88.9	MEG+HD-EEG (resection ref.)
2	Guo et al. (2022)7	75.3	93.2	76.5	66.7	100.0	44.4	PET+MRI+MEG (resection ref.)
6	Widjaja et al. (2013)26	N/A	N/A	85.0	99.1	95.0 (MEG or PET)	100.0 (MEG and PET)	PET+MRI combinations (MEG or PET, MEG and PET)
7	Schneider et al. (2012)8	N/A	N/A	70.0	50.0	70.0	87.5	iEEG comparison
8	Knowlton et al. (2008)27
	MEG or PET†	N/A	N/A	64.3	78.6	80.0	40.0	PET/SPECT+MEG combinations (MEG or PET/SPECT, MEG and PET/SPECT)
	MEG and PET*	N/A	N/A			15.6	86.7
	MEG or SPECT†	N/A	N/A	60.0	87.5	80.0	40.0
	MEG and SPECT*	N/A	N/A			24.0	90.0
9	Pataraia et al. (2004)17	57.3	40.0			N/A	N/A	Compared to V-EEG only (not surgical)

Guo et al. (2022)7: Engel class I outcome was 71.2% with MEG alone and 94.5% when MEG was combined with PET and MRI, indicating a 23.3% improvement with multimodal integration. Pataraia et al. (2004)17: MEG improved localization compared to scalp EEG in 26 out of 65 patients (40.0%), as defined by resection overlap and electrographic concordance. The study was defined as cases where MEG, performed after scalp V-EEG, provided additional localizing information that more accurately matched the resection zone. MEG localization was fully concordant with resection in 57.3% of patients (47/82). MEG improved localization over V-EEG in 40.0% of patients (26/65).

MEG, magnetoencephalography; Sn, sensitivit; Sp, specificity; HD-EEG, high-density electroencephalography; SOZ, seizure onset zone; ref., reference; PET, positron emission tomography; MRI, magnetic resonance imaging; N/A, not applicable; iEEG, intracranial electroencephalography; SPECT, single photon emission computed tomography; V-EEG, video electroencephalography; EEG, electroencephalography.

^*AND: outcome considered favorable only if both MEG and PET localized concordantly.

^†OR: outcome considered favorable if either MEG or PET localized to the resection site.

Similarly, Guo et al.7 reported that the combined use of PET, MRI, and MEG resulted in the highest concordance with the surgical resection site (94.5%), significantly outperforming MEG alone (71.2%) or PET combined with MRI (82.2%). In Pataraia et al.,17 MEG showed complete concordance with the resection zone in 57.3% (47/82) of patients. Additionally, in a subgroup of 65 patients, MEG improved localization compared to scalp video-EEG in 40.0% (26/65), highlighting its complementary role in surgical planning.17

Surgical outcomes and odds ratios (ORs)

Five studies provided data on surgical outcomes based on Engel class I seizure freedom rates (Table 2). Although MEG alone yielded Engel class I outcomes ranging from 46.5% to 68.8%, its combination with HD-EEG, PET, or iEEG significantly improved seizure-free rates, with some studies reporting rates above 90%.

Table 2.

Surgical outcomes (Engel class I) following MEG-based presurgical evaluation

No.	Study	MEG alone Engel I	MEG+other Engel I	Comparator	Notes
1	Chikara et al. (2023)16	46.5	87.0	HD-EEG	Significant improvement with HD-EEG
3	Tripathi et al. (2021)18	68.8 (MEG-guided group)	58.7 (HD-EEG-guided group)	V-EEG, HD-EEG, PET	Multiple combinations evaluated
4	Wang et al. (2019)19	30.0	68.2	MAP	MAP (including MEG, PET, and SISCOM)
5	Zhang et al. (2016)20	68.9	79.3	MEG+PET	-MEG+PET vs. MEG comparision -Subgroup (unilobar, nultilobar)
	Unilobar	78.6	91.3
	Multilobar	60.6	33.3
6	Widjaja et al. (2013)26	60.6	60.6	FDG-PET	Same for both

Tripathi et al. (2021)18: MEG-guided group: when MEG was considered in surgical planning; HD-EEG-guided group: when HD-EEG was considered in surgical planning.

MEG, magnetoencephalography; HD-EEG, high-density electroencephalography; V-EEG, video electroencephalography; PET, positron emission tomography; MAP, morphometric analysis program; SISCOM, subtraction ictal SPECT co-registered to MRI; FDG-PET, 18F-fluorodeoxyglucose positron emission tomography; SPECT, single-photon emission computed tomography; MRI, magnetic resonance imaging.

Table 3 OR data from three studies evaluating the effect of MEG on surgical outcomes. Chikara et al.16 found that combining MEG with HD-EEG resulted in an Engel class I outcome in 87.0% of patients, which was a significant improvement compared to MEG alone (46.5%) or HD-EEG alone (61%), yielding an OR of 1.33 (95% confidence interval [CI], 1.07–1.83; p=0.01) for the combined approach over individual modalities. Tripathi et al.18 reported a seizure-free rate of 68.8% when MEG was considered in surgical planning, versus 18.7% for HD-EEG, with an OR of 2.35 (95% CI, 0.68–7.86; p=0.18). Schneider et al.8 reported a significant effect of MEG, with an OR of 16.3 (95% CI, 1.35–197.77; p=0.02), based on a model comparing patients with and without MEG-guided planning, even when iEEG was used in both groups.

Table 3.

Odds ratio comparisons across studies

No.	Study	Comparison type	Engel I outcome	Odds ratio‡ (95% confidence interval)	p-value	Notes
1	Chikara et al. (2023)16	MEG+HD-EEG	87.0	1.33 (1.07–1.83)	0.01	Significant benefit with HD-EEG addition
		MEG	46.5	1.01 (0.99–1.05)†	0.15
3	Tripathi et al. (2021)18	MEG-guided	68.8	2.35 (0.68–7.86)	0.18	Multiple modalities evaluated; MEG-guided strategy
		Not MEG-guided	18.7	N/A	N/A
7	Schneider et al. (2012)8	MEG+iEEG	N/A	16.3 (1.35–197.77)*	0.02	Improved outcomes with MEG+iEEG combination
		MEG	N/A	2.3 (0.34–16.18)	0.26

Odds ratio values compare seizure freedom rates between groups with and without MEG or MEG-based evaluation.

MEG, magnetoencephalography; HD-EEG, high-density electroencephalography; N/A, not applicable; iEEG, intracranial electroencephalography.

^*The odds ratio of 16.3 reported in Schneider et al. (2012)8 reflects the added value of MEG in surgical planning, as evaluated in a statistical model comparing patients with and without MEG-guided assessment.

^†MEG-guided: surgical decision-making explicitly influenced by MEG results (e.g., guiding electrode implantation or resection strategy). MEG alone indicates odds ratios calculated solely from MEG as a diagnostic test result without its incorporation into surgical planning.

^‡In Tripathi et al. (2021),18 the comparison was between MEG-guided (odds ratio, 2.35) and HD-EEG-guided (odds ratio, 1.36) surgical planning; confidence interval was not reported for HD-EEG.

Wang et al.19 investigated various combinations of non-invasive examinations in 22 MRI-negative patients. While individual modalities showed varying success rates (MEG, 30.0%), the combination of morphometric analysis program, MEG, and subtraction Ictal SPECT co-registered to MRI yielded the highest localization success rate at 68.2%. This suggests that multiple tests, when combined, collectively enhance localization accuracy.19

Zhang et al.20 compared the localization success rates of MEG and PET, both individually and in combination, in 85 MRI-negative patients. MEG alone achieved a success rate of 68.9%, PET alone was 56.6%, and their combination significantly improved the rate to 79.3%. Subgroup analysis further revealed that for patients with unilobar lesions, the MEG+PET combination showed a particularly high success rate of 91.3%. However, for multi-lobar epilepsy, MEG as a standalone examination exhibited the highest success rate at 60.6%, suggesting that the optimal strategy may vary with epilepsy type.20

These findings collectively support MEG’s value not only in localizing epileptogenic zones but also in improving surgical outcomes, particularly when used in conjunction with other neuroimaging or electrophysiological modalities.

Subgroup analysis

Subgroup analysis based on patient age, epilepsy type, and MRI status is summarized in Fig. 3 and Table 4. MEG was particularly effective in pediatric patients and MRI-negative cases, often achieving localization and seizure-free rates exceeding 80% when used in strategies. Temporal lobe epilepsy (TLE) cases consistently showed high localization concordance and favorable surgical outcomes with MEG integration. However, in cases of multi-lobar epilepsy, MEG as a standalone examination sometimes demonstrated the highest localization success rate. This underscores the importance of considering epilepsy subtype and tailoring examination strategies, which may include MEG alone or in appropriate combinations with other modalities.

Figure 3.

Comparative effectiveness of MEG across subgroups. Pediatric: Engel I 87% from Chikara et al.16 Adult: combined estimates from multiple studies. MRI-negative: Zhang et al.,20 localization ~75%. MRI-positive: Zhang et al.,20 refinement role. TLE: Pataraia et al.,17 localization 77%; ETLE: Pataraia et al.,17 localization ~60%. MRI, magnetic resonance imaging; TLE, temporal lobe epilepsy; ETLE, extra-temporal lobe epilepsy; MEG, magnetoencephalography.

Table 4.

Subgroup summary of MEG effectiveness

Subgroup	Localization rate	Engel class I	Notable findings
Pediatric	75–87	67–87	Best for FCD; MRI-negative cases; MEG+PET/HD-EEG
Adult	65–77	60–70	TLE high concordance with PET/MRI
MRI-negative	>70	High	Added new localization; key in surgical planning
MRI-positive	Moderate	Moderate	Confirmed or refined ambiguous zones
TLE	~77	High	Strong localization; favorable outcomes
ETLE	~60	Moderate	More variable; needs multimodal integration

MEG, magnetoencephalography; FCD, focal cortical dysplasia; MRI, magnetic resonance imaging; PET, positron emission tomography; HD-EEG, high-density electroencephalography; TLE, temporal lobe epilepsy; ETLE, extra-temporal lobe epilepsy.

TLE cases exhibited high localization concordance, whereas ETLE showed greater variability and benefited more from integration. MEG’s additive value was especially clear in MRI-negative populations, where it frequently provided new localizing information critical to surgical planning. These patterns highlight the importance of individualized strategies in presurgical evaluation.

Conclusion

The current systematic review confirms that MEG plays a significant role in the presurgical evaluation of epilepsy.16 Our findings demonstrate that MEG, particularly when integrated within assessment strategies, consistently enhances localization accuracy and improves post-surgical seizure freedom.7,8,16–21 As a non-invasive tool, MEG offers a crucial advantage by potentially reducing the need for more invasive procedures like iEEG, thereby minimizing patient risks and discomfort, while still providing critical localization information.8

This systematic review’s primary finding reinforces the pivotal role of MEG not as a standalone tool, but as an indispensable component of presurgical evaluation. All studies examining the combination of MEG with non-invasive EEG, PET, and SPECT confirmed that each modality contributes unique advantages and, when used in conjunction, can relatively improve localization accuracy.7,8,16–21 This synergistic effect allows for a more comprehensive understanding of the epileptogenic zone, mitigating the limitations of any single imaging modality and ultimately leading to more precise surgical targets. In patients with non-localizing or discordant EEG/MRI results, MEG often provides critical complementary localization data that, when combined with other examinations, can significantly influence surgical decision-making and reduce false positives and false negatives.21 For patients, this translates into a higher probability of seizure freedom post-surgery, significantly improving their quality of life and reducing long-term disability.16,18,20

Subgroup analyses by age and epilepsy subtype revealed differential effectiveness of MEG, highlighting tailored strategies for optimal outcomes. Pediatric patients, especially those with focal cortical dysplasia (FCD) or MRI-negative epilepsy, consistently derived the greatest benefit from MEG. For instance, combining HD-EEG with MEG in pediatric FCD cases yielded high Engel class I outcomes of up to 87% and superior diagnostic accuracy for surgical resection site prediction.16 This underscores MEG’s high sensitivity in detecting subtle cortical abnormalities in developing brains and its value in reducing the need for invasive procedures in this vulnerable population.

In adult patients, while TLE cases consistently showed high localization concordance and favorable surgical outcomes with MEG integration, 7,17 ETLE showed greater variability and often benefited more from integration. MEG’s additive value was particularly clear in MRI-negative populations, where it frequently provided new localizing information critical to surgical planning.21 The findings also offered nuanced insights into lesion complexity: while for unilobar lesions, combining MEG with PET resulted in remarkably high success rates (e.g., 91.3% with MEG+PET for unilobar lesions), for multi-lobar epilepsy, MEG as a standalone examination sometimes demonstrated the highest localization success rate (e.g., 60.6% with MEG alone for multi-lobar lesions).20 This underscores the importance of considering epilepsy subtype and tailoring examination strategies, which may include MEG alone or in appropriate combinations with other modalities.

These findings are in line with the detailed analysis by Zhang et al.,20 who reported that PET and PET+MEG achieved significantly higher localization rates in unilobar than in multilobar epilepsy (PET: 74.1% vs. 34.6%; PET+MEG: 91.3% vs. 33.3%), whereas MEG only did not show a statistically significant difference between the two groups (78.6% vs. 60.6%). Although these PET and PET+MEG values were reported in the original study and are not included in our table, they are presented here for contextual comparison. The authors explained that PET signals in multilobar epilepsy tend to disperse across multiple lobes, reducing sensitivity, while MEG can capture more widespread cortical activity, thereby maintaining relatively preserved performance. However, they also emphasized that discordant or even contradictory findings between PET and MEG were frequent in multilobar cases, highlighting the continued need for invasive monitoring in such patients.

While the initial uptake of MEG examinations in 2023 showed promising numbers for a newly introduced service (162 cases), the most recent health insurance review and assessment service statistics indicate a slight decline in 2024 (156 cases).22 This modest reduction is unlikely to reflect diminished clinical demand; rather, it points to persistent barriers such as high cost, limited device availability, and geographic constraints. Supporting this interpretation, the Korean Epilepsy Society has estimated that approximately 2,500 patients require MEG annually, while Professor Seung-Chyul Hong reported that although the number of patients requiring epilepsy surgery has been increasing by approximately 1,000 each year, fewer than 300 surgeries are actually performed annually in Korea.23,24 These figures underscore the gap between potential demand and actual access, highlighting the need for policy interventions to ensure equitable patient care.

Furthermore, the current geographical concentration of the sole MEG device in the metropolitan area significantly contributes to regional disparity in access to advanced epilepsy care, as evidenced by the high proportion of consultations from the Seoul area through the National Epilepsy Care Center Helpline.9,11–13 To mitigate these disparities and ensure equitable access, national-level intervention is imperative. As advanced and high-cost medical equipment, MEG systems require substantial investment for both acquisition and maintenance, posing significant barriers for individual institutions.13 Therefore, advocating for comprehensive national support for epilepsy care, including the strategic expansion and regional distribution of MEG facilities, is crucial.9 This aligns with efforts in other developed countries that provide robust national support for essential epilepsy infrastructure and services. Such policy initiatives would not only enhance clinical outcomes but also significantly improve the quality of life for a broader population of epilepsy patients across the country.

This review has several limitations. First, the included studies exhibited considerable heterogeneity in terms of patient populations, MEG acquisition protocols, and outcome measures, which limited our ability to perform a meta-analysis. Second, most studies were observational and conducted in high-volume tertiary care centers, potentially reducing the generalizability of the findings to community hospital settings. Third, the risk of publication bias cannot be ruled out, as studies with negative or inconclusive MEG results may be underrepresented in the literature. Finally, due to the narrative synthesis approach, the strength of the associations reported should be interpreted with caution.

Although this review did not include a formal economic evaluation, findings from Canadian model-based studies suggest that incorporating MEG-particularly within combination strategies-may increase initial evaluation costs but ultimately yield substantial long-term benefits, including improved quality-adjusted life years and higher rates of seizure freedom and surgical success.25

From a patient-centered perspective, these long-term benefits, such as reduced seizures, improved quality of life, and decreased lifelong medical burden, far outweigh the initial investment, underscoring the broader societal value of the technology. Despite these promising findings, a critical need remains for further cost-effectiveness analyses that consider the specific healthcare context and economic factors of South Korea, as well as for large-scale, prospective local studies to fully evaluate its real-world impact and cost-effectiveness within the national healthcare system.

Future research should focus on standardizing MEG protocols and conducting high-quality, prospective research to enhance reproducibility and validate its predictive value for surgical outcomes across diverse epilepsy subtypes. Furthermore, future studies should also focus on implementation science to identify optimal strategies for integrating MEG into regional epilepsy care pathways in countries with limited resources, addressing issues of accessibility, cost-effectiveness, and training for local specialists.

MEG is a powerful tool in the presurgical evaluation of epilepsy. It provides unique and complementary information to traditional imaging and EEG, particularly in complex cases such as MRI-negative or extratemporal epilepsy. When integrated within an assessment strategy, MEG significantly enhances localization accuracy and surgical planning.

For South Korea, optimizing the clinical impact of MEG necessitates strategic national support for its expanded regional accessibility and standardized utilization.9 This review highlights the clear clinical benefits of MEG when used in strategies, but also underscores the significant disparities in access due to the limited number of devices currently available nationwide. Efforts to increase nationwide accessibility, promote its effective integration into epilepsy care pathways, and ensure equitable distribution of this valuable resource will be essential for maximizing its benefits for epilepsy patients. Such policy-driven initiatives are crucial for enhancing patient outcomes and improving the overall quality of epilepsy care in the country.