EPICURUS: E-field-based spatial filtering procedure for an accurate estimation of local EEG activity evoked by Transcranial Magnetic Stimulation

EPICURUS: E-field-based spatial filtering procedure for an accurate estimation of local EEG activity evoked by Transcranial Magnetic Stimulation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Short Report EPICURUS: E-field-based spatial filtering procedure for an accurate estimation of local EEG activity evoked by Transcranial Magnetic Stimulation Xavier Corominas-Teruel, Tuomas P. Mutanen, Carlo Leto, Maria Teresa Colomina, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7593453/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background : The concurrent use of Transcranial magnetic stimulation (TMS) with electroencephalography (EEG) is increasingly integrated into research and clinical protocols. However, a reliable isolation of EEG responses that are locally evoked by TMS at the targeted cortical sites remains challenging. Methods : We introduce EPICURUS, a novel spatial filtering approach for TMS-EEG that uses individualized MRI-based simulations of the TMS-induced electric field (E-field) to define the spatial extent of locally evoked activity. This method guides the reconstruction of EEG signals originating from the primary stimulation site while minimizing crosstalk from distant, non-targeted sources. Results : Application of EPICURUS revealed a marked reduction in late transcranial evoked potential (TEP) components (>100 ms), indicating that E-field-based spatial filtering effectively suppresses contributions from indirect or multisensory sources, particularly when combined with auditory suppression techniques. Conclusion : By leveraging the spatial precision of individualized E-field modeling, EPICURUS enhances the specificity of EEG signal reconstruction, offering a promising tool for improving the spatiotemporal resolution of cortical responses directly evoked by TMS. Brain Stimulation Electrical field simulation Transcranial magnetic stimulation concurrent TMS-EEG spatial filtering Figures Figure 1 Figure 2 HIGHLIGHTS The lack of spatial specificity of EEG prevents an accurate estimation of the locally activated EEG sources, hindering the interpretation of TMS effects. We here developed a spatial filtering approach for TMS-EEG based on individual estimation of E-field spread following stimulation. Our E-field spatial filter, combined with auditory constraints, reduced non-relevant distant signal components and improved the reliability of local EEG estimates. INTRODUCTION Electroencephalography (EEG) is a neurophysiological approach increasingly employed to monitor the impact of Transcranial Magnetic Stimulation (TMS) on brain activity. However, the activity recorded by such a high-temporal resolution technique after a TMS pulse remains spatially unspecific and reflects a combination of mixed cerebral sources directly or indirectly activated by TMS. This phenomenon occurs because, besides relevant local direct effects on the targeted cortical site, TMS engages activity in interconnected distant areas and networks [ 1 , 2 ], while also activating non-relevant regions due to TMS-generated sensory input in the auditory, tactile, and vestibular domains [ 3 – 7 ]. Here, we present an innovative approach to isolate activity locally elicited by TMS pulses within the targeted cortical region, defined by the modeled electric field extension, while minimizing unspecific contributions from non-relevant (directly and indirectly) activated sources. The method (EPICURUS; E -field based s P atial f I ltering for C oncurrent EEG-TMS data U sed to ta R get so U rce S ignal) has been designed as a spatial filter for TMS-EEG datasets and combines E-field simulations [ 8 ] with a Minimum Norm-like EEG source Estimator (MNE), minimizing signal leakage and crosstalk from distal brain locations [ 9 – 11 ]. MATERIALS AND METHODS Participants To test EPICURUS performance, we used it to analyze a TMS-EEG dataset of n = 18 healthy participants (10 females, 8 males, 26 ± 5 years old), all right-handed, meeting safety criteria for MRI [ 12 ] and TMS [ 13 ], who provided written consent to participate in the study. Our protocol followed the Declaration of Helsinki and received approval from a local ethical committee ( Comité de Protection des Personnes, Ile de France I ). TMS-EEG protocol and data processing Participants received 80 single TMS pulses (with a variable inter-stimulus interval of 5-7sec) over the left primary motor cortex (M1) while we continuously recorded scalp EEG (25kHz sampling rate; ActiChamp, with 65 electrodes placed following the International 10/10 system coordinates). The left M1 'hotspot’ was defined as the optimal cortical location for eliciting Motor-Evoked Potentials in the First Dorsal Interosseous muscle at rest [ 14 ]. TMS was administered with a D70 mm figure-of-eight alpha coil attached to a biphasic stimulator (Magstim Rapid2) at a fixed intensity (60% of the maximal stimulator output; MSO). For each participant, the TMS coil position was maintained consistently throughout the experiment using an MRI-frameless stereotactic system (Brainsight, Rogue Solutions Inc., Montréal, QC, Canada). The TMS-EEG dataset was processed using the MATLAB-based toolbox EEGLAB [ 15 ] and its TMS–EEG signal analyzer plugin (TESA [ 16 ]). Data was epoched around the TMS pulse (from − 1200 to 1200ms – w.r.t. – the TMS pulse) and baseline corrected (from − 500 to -10ms). TMS-pulse electromagnetic (from − 2ms to 12ms w.r.t the TMS pulse) and recharging artifacts (visually identified [ 17 ]) were removed and interpolated using a spline function. The data were downsampled to 1 kHz, and the most artifacted channels were temporarily removed from analyses, and the independent component analysis (ICA) was employed to remove ocular artifacts. Next, the SOUND algorithm was used for noise suppression, replacement of removed channels, and signal re-referencing to a common average [ 18 ]. The SSP-SIR algorithm was then used to suppress TMS-evoked muscle artifacts [ 19 ], and a second ICA was computed to remove residual continuous muscle artifacts. Finally, the data were bandpass filtered with a fourth-order Butterworth filter (2-45Hz) and baseline corrected. E-field-based spatial filter Our pipeline integrates a previously developed spatial filter approach based on cross-talk functions (DeFleCT; [ 9 ]) with E-field distribution modeled using SimNIBS 4.0 [ 8 ]. First, we reconstructed individual finite element head models (FEM [ 20 ]) for each participant and computed the leadfield matrix employing the Helsinki BEM framework from the FEM model [ 10 ]. We simulated the E-field distribution induced by TMS at the intensity of 60% MSO with SimNIBS (version 4.0, 68,8dI/dtmax = 60% MSO; [ 21 ]), using TMS coil positions recorded during the experimental session. We employed each participant’s E-field spatial distribution to identify the cortical area most impacted by TMS (> 80% E-field peak strength; region of interest–ROI). Next, a spatial filter with distributed source minimization constraints (i.e., crosstalk from outside the established ROI) was built for each participant. The spatial filtering strategy mirrors Special Case 2 of the DeFleCT implementation [ 9 ], using an L2 minimum-norm estimate with a normalization constraint tailored to the TMS targeted sources combined with a noise minimization constraint (Fig. 1 ). To validate our approach and explore the efficacy of the filter, we implemented an additional filtering procedure; a discrete source constraint minimizing signal components arising from sources in the auditory cortex (~ Brodman areas 22,41, and 42; [ 22 ]; bilaterally delineated on each individual head model). Hence, the original EEG signals were filtered to generate two dataset conditions: one using the E-field-based filter alone and another adding to the former the above-mentioned auditory constraints. Spatial filter validation and statistical analysis To validate our spatial filtering pipeline, we first identified TEPs from all available EEG sensors (Fig. 2 .A.1, A.2) and characterized their latency, morphology, and topographies. Next, we analyzed TEPs from a cluster of electrodes in the vicinity of the TMS stimulation site (i.e., mean signal from C1, C3, CP1, and CP3; sensor-space original signal; Fig. 2 .A.3). Additionally, we employed a classical MNE estimator, without noise minimization (i.e., source-level original signal), to reconstruct EEG time-series directly from the sources impacted by the stimulation (i.e., the sources identified within the E-field-simulation-based ROI; Fig. 2 .A.3). The sensor-level and source-level original TEPs served as reference data conditions for our analysis and were statistically compared to the spatially filtered signals (E-field-based and E-field-based with auditory constraints). Comparisons were performed with non-parametric cluster-based permutation statistics employing one-way 1x3 ANOVA (FieldTrip; [ 23 , 24 ]; 50000 iterations, α = 0.05, p < 0.05, time-window: from 0ms to 400ms -w.r.t. TMS pulse). Whenever needed, post-hoc tests were conducted using cluster-based non-parametric statistics with two-tailed dependent-samples t-tests (50000 iterations, α = 0.05, p < 0.05), employing Bonferroni correction for multiple comparisons. RESULTS Comparisons between E-field-based filtered datasets and the original source-level and sensor-level signals revealed significant differences, identifying a significant cluster ~ 175ms following TMS pulse ( F (2, 32) = 9.1, p = 0.002; Fig. 2 .B, left panel). Post-hoc comparisons indicated significant differences between the source-level and the sensor-level original signals (paired t-test: t = 3.6, p corrected = 0.006), and between the E-field-based filtered signal and sensor-level original signal (paired t-test: t = 2.9, p corrected = 0.04). However, no significant differences were observed between the E-field-based filtered signal and the original source-level dataset (paired t-test: t = 2.3, p corrected = 0.3). Comparisons between E-field-based filtered signals with auditory constraints, the original source-level and sensor-level signals revealed significant differences across signals and revealed two clusters with different latencies (cluster 1, ~ 115ms, F (2, 32) = 7.7, p = 0.03; cluster 2, ~ 175ms, F (2, 32) = 9.7, p = 0.001; Fig. 2 .B, central panel). Post-hoc comparisons indicated significant differences between the E-field-based filtered with auditory constraints and the sensor level original signal on both clusters (cluster 1, paired t-test: ~115ms, t = -4, p corrected = 0.01 / cluster 2, paired t-test: ~175ms, t = 3.2, p corrected = 0.01), and between the E-field-based filtered with auditory constraints and the source level original signal (paired t-test: ~115ms, t = -2.7, p corrected = 0.05). Finally, the comparison between the E-field-based filtered signal and the E-field-based with auditory constraints conditions revealed significant differences in the same two late latency clusters described in the previous analysis (cluster 1, paired t-test: ~115ms, t = -4.5, p corrected = 0.0004; cluster 2, paired t-test: ~175ms, t = 4.6, p corrected = 0.001; Fig. 2 .B, right panel). DISCUSSION We designed and implemented EPICURUS, a spatial filtering approach for TMS-EEG data that employs the TMS coil position during the experimental session, the E-field distribution biophysical model, and an MNE source reconstruction method minimizing signal leakage and crosstalk from distant brain sources. Our results revealed three key findings. First, the shape and latency of E-field-based filtered TEPs (with and without auditory constraints) were comparable to those originally recorded or estimated (source and sensor level), supporting the reliability and robustness of an approach that does not alter the overall time course of TMS-evoked signals. Second, E-field-based filtered TEPs (with and without auditory constraints) showed lower amplitude than the sensor-level original TEPs, especially at later components (> 100 ms post TMS-pulse). However, only TEPs processed with EPICURUS including auditory constraints reached statistically significant differences compared to original source-level TEPs. Third, signals reconstructed with the E-field-based filtering method differed from those reconstructed with E-field-based filtering combined with auditory constraints, particularly ~ 115ms. Prior work has shown that late TEP components, such as those present in our data at ~ 115 and ~ 175 ms (N115 and P175) with fronto-central topographies (Fig. 2 .A.2), typically reflect the engagement of distant brain regions indirectly activated by TMS (i.e., multisensory responses [ 3 – 5 , 7 ]). We here hypothesize that statistical differences between filtered and original signals reported may arise from the attenuation of distant sources, which, at least in part, reflect the indirect multisensory EEG responses to TMS. A likely direct link to auditory evoked activity is supported by our findings showing that the addition of auditory constraints significantly reduces the amplitudes of such late TEP components. Although our filtering approach efficiently suppressed the contribution of TMS activity associated with multisensory responses, the primary aim of EPICURUS is to increase the reliability of local evoked responses by minimizing the distant non-relevant sources. The suppression of multisensory responses in our TMS-EEG example is an indirect consequence of this aim. To effectively minimize multisensory responses, proper experimental procedures during experimental sessions (e.g., noise masking, control conditions), potentially combined with other analysis strategies, will be necessary [ 25 – 27 ]. Declarations COMPETING INTERESTS The authors declare no competing financial interests. DATA AND CODE AVAILABILITY The datasets produced and/or analyzed during the current study cannot be made publicly available due to institutional regulations. However, they can be obtained from the corresponding and/or first authors of the manuscript upon reasonable request. MATLAB codes for the pre-processing and filtering implementation can be found at the permanent link: https://github.com/XavierCorominas/Source-Based-filters-for-brain-stimulation-readouts . CREDIT AUTHORSHIP CONTRIBUTION STATEMENT Conceptualization: XC-T, TM, MB, AV-C. Methodology: XC-T, TM, MB. Data curation and analysis: XC-T. Funding acquisition: CG, MB, AV-C. Supervision: MB, AV-C. Writing – original draft: XC-T, AV-C & MB. Writing – review and editing: XC-T, TM, CL, CG, MT-C, MB, AV-C. ACKNOWLEDGEMENTS This work was supported by the Agence Nationale de la Recherche in France (BrainMAG; ANR-19-CE37-0021; AV-C), the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement 897941 (MB), the Big Brain Theory (BBT-3; FORTE) internal ICM grant (CG, AV-C & MB), ICM CARNOT training grant (MB, CG), IHU-ICM CARNOT Maturation grant, and additional funding came from Investissements d’avenir (ANR-10- IAIHU-0006) awarded to XC-T and AV-C for associated projects. References Bortoletto M, Veniero D, Thut G, Miniussi C (2015) The contribution of TMS–EEG coregistration in the exploration of the human cortical connectome. Neurosci Biobehav Rev 49:114–124. https://doi.org/https://doi.org/10.1016/j.neubiorev.2014.12.014 Bracco M, Mutanen TP, Veniero D, Thut G, Robertson EM (2025) Protocol to assess changes in brain network resistance to perturbation during offline processing using TMS-EEG. STAR Protoc 6:103622. https://doi.org/https://doi.org/10.1016/j.xpro.2025.103622 Rocchi L, Di Santo A, Brown K, Ibáñez J, Casula E, Rawji V et al (2021) Disentangling EEG responses to TMS due to cortical and peripheral activations. Brain Stimulation: Basic Translational Clin Res Neuromodulation 14:4–18. https://doi.org/10.1016/j.brs.2020.10.011 Conde V, Tomasevic L, Akopian I, Stanek K, Saturnino GB, Thielscher A et al (2019) The non-transcranial TMS-evoked potential is an inherent source of ambiguity in TMS-EEG studies. NeuroImage 185:300–312. https://doi.org/https://doi.org/10.1016/j.neuroimage.2018.10.052 Gordon PC, Desideri D, Belardinelli P, Zrenner C, Ziemann U (2018) Comparison of cortical EEG responses to realistic sham versus real TMS of human motor cortex. Brain Stimulation: Basic Translational Clin Res Neuromodulation 11:1322–1330. https://doi.org/10.1016/j.brs.2018.08.003 Hernandez-Pavon JC, Veniero D, Bergmann TO, Belardinelli P, Bortoletto M, Casarotto S et al (2023) TMS combined with EEG: Recommendations and open issues for data collection and analysis. Brain Stimul 16:567–593. https://doi.org/https://doi.org/10.1016/j.brs.2023.02.009 Niessen E, Bracco M, Mutanen TP, Robertson EM (2021) An analytical approach to identify indirect multisensory cortical activations elicited by TMS? Brain Stimulation: Basic, Translational, and Clinical Research in Neuromodulation. 14(2):376–378. https://doi.org/10.1016/j.brs.2021.02.003 Thielscher A, Antunes A, Saturnino GB (2015) Field modeling for transcranial magnetic stimulation: A useful tool to understand the physiological effects of TMS? 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 222–5. https://doi.org/10.1109/EMBC.2015.7318340 Hauk O, Stenroos M (2014) A framework for the design of flexible cross-talk functions for spatial filtering of EEG/MEG data: DeFleCT. Hum Brain Mapp 35:1642–1653. https://doi.org/https://doi.org/10.1002/hbm.22279 Stenroos M, Hunold A, Haueisen J (2014) Comparison of three-shell and simplified volume conductor models in magnetoencephalography. NeuroImage 94:337–348. https://doi.org/https://doi.org/ 10.1016/j.neuroimage.2014.01.006 Hämäläinen MS, Ilmoniemi RJ (1994) Interpreting magnetic fields of the brain: minimum norm estimates. Med Biol Eng Comput 32:35–42. https://doi.org/10.1007/BF02512476 Shellock FG, Spinazzi A, MRI Safety Update (2008) : Part 2, Screening Patients for MRI. American Journal of Roentgenology 2008;191:1140–9. https://doi.org/10.2214/AJR.08.1038.2 Rossi S, Hallett M, Rossini PM, Pascual-Leone A (2009) Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clin Neurophysiol 120:2008–2039. https://doi.org/https://doi.org/10.1016/j.clinph.2009.08.016 Chen R, Cros D, Curra A, Di Lazzaro V, Lefaucheur J-P, Magistris MR et al (2008) The clinical diagnostic utility of transcranial magnetic stimulation: Report of an IFCN committee. Clin Neurophysiol 119:504–532. https://doi.org/https://doi.org/10.1016/j.clinph.2007.10.014 Delorme A, Makeig S (2004) EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods 134:9–21. https://doi.org/https://doi.org/10.1016/j.jneumeth.2003.10.009 Rogasch NC, Sullivan C, Thomson RH, Rose NS, Bailey NW, Fitzgerald PB et al (2017) Analysing concurrent transcranial magnetic stimulation and electroencephalographic data: A review and introduction to the open-source TESA software. NeuroImage 147:934–951. https://doi.org/https://doi.org/10.1016/j.neuroimage.2016.10.031 Veniero D, Bortoletto M, Miniussi C (2009) TMS-EEG co-registration: On TMS-induced artifact. Clin Neurophysiol 120:1392–1399. https://doi.org/https://doi.org/ 10.1016/j.clinph.2009.04.023 Mutanen TP, Metsomaa J, Liljander S, Ilmoniemi RJ (2018) Automatic and robust noise suppression in EEG and MEG: The SOUND algorithm. NeuroImage 166:135–151. https://doi.org/https://doi.org/10.1016/j.neuroimage.2017.10.021 Mutanen TP, Kukkonen M, Nieminen JO, Stenroos M, Sarvas J, Ilmoniemi RJ (2016) Recovering TMS-evoked EEG responses masked by muscle artifacts. NeuroImage 139:157–166. https://doi.org/https://doi.org/10.1016/j.neuroimage.2016.05.028 Puonti O, Van Leemput K, Saturnino GB, Siebner HR, Madsen KH, Thielscher A (2020) Accurate and robustwhole-head segmentation from magnetic resonance images for individualized head modeling. NeuroImage 219:117044. https://doi.org/https://doi.org/10.1016/j.neuroimage.2020.117044 Drakaki M, Mathiesen C, Siebner HR, Madsen K, Thielscher A (2022) Database of 25 validated coil models for electric field simulations for TMS. Brain Stimul 15:697–706. https://doi.org/https://doi.org/10.1016/j.brs.2022.04.017 Hackett TA (2015) Chapter 2 - Anatomic organization of the auditory cortex. In: Aminoff MJ, Boller F, Swaab DF, editors. Handb Clin Neurol, vol. 129, Elsevier; pp. 27–53. https://doi.org/https://doi.org/10.1016/B978-0-444-62630-1.00002-0 Oostenveld R, Fries P, Maris E, Schoffelen J-M, FieldTrip (2011) Open Source Software for Advanced Analysis of MEG, EEG, and Invasive Electrophysiological Data. Comput Intell Neurosci 2011:156869. https://doi.org/10.1155/2011/156869 Maris E, Oostenveld R (2007) Nonparametric statistical testing of EEG- and MEG-data. J Neurosci Methods 164:177–190. https://doi.org/https://doi.org/ 10.1016/j.jneumeth.2007.03.024 Metsomaa J, Song Y, Mutanen TP, Gordon PC, Ziemann U, Zrenner C et al Adapted Beamforming: A Robust and Flexible Approach for Removing Various Types of Artifacts from TMS–EEG Data. Brain Topogr 2024. https://doi.org/10.1007/s10548-024-01044-4 Hernandez-Pavon JC, Kugiumtzis D, Zrenner C, Kimiskidis VK, Metsomaa J (2022) Removing artifacts from TMS-evoked EEG: A methods review and a unifying theoretical framework. J Neurosci Methods 376:109591. https://doi.org/https://doi.org/10.1016/j.jneumeth.2022.109591 Biabani M, Fornito A, Mutanen TP, Morrow J, Rogasch NC (2019) Characterizing and minimizing the contribution of sensory inputs to TMS-evoked potentials. Brain Stimul 12:1537–1552. https://doi.org/https://doi.org/10.1016/j.brs.2019.07.009 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7593453","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Short Report","associatedPublications":[],"authors":[{"id":518836271,"identity":"9f54d740-a1a5-41bf-b7ca-462df7a24c7c","order_by":0,"name":"Xavier Corominas-Teruel","email":"","orcid":"","institution":"APHP-Hôpital de la Pitié Salpêtrière, FRONTLAB Team","correspondingAuthor":false,"prefix":"","firstName":"Xavier","middleName":"","lastName":"Corominas-Teruel","suffix":""},{"id":518836272,"identity":"aceee7b9-fd61-46e0-bb49-a9cf2b8b6a6f","order_by":1,"name":"Tuomas P. Mutanen","email":"","orcid":"","institution":"Aalto University School of Science","correspondingAuthor":false,"prefix":"","firstName":"Tuomas","middleName":"P.","lastName":"Mutanen","suffix":""},{"id":518836273,"identity":"ebc2b82b-61c0-47a4-8e56-7a6ff88531fe","order_by":2,"name":"Carlo Leto","email":"","orcid":"","institution":"APHP-Hôpital de la Pitié Salpêtrière, FRONTLAB Team","correspondingAuthor":false,"prefix":"","firstName":"Carlo","middleName":"","lastName":"Leto","suffix":""},{"id":518836274,"identity":"cc78e745-27c2-4bb1-ab3b-3e30e99d9e5c","order_by":3,"name":"Maria Teresa Colomina","email":"","orcid":"","institution":"Universitat Rovira i Virgili, NEUROLAB","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Teresa","lastName":"Colomina","suffix":""},{"id":518836275,"identity":"47be6fcc-b9ad-4aa9-8565-3f0786afe55a","order_by":4,"name":"Cécile Gallea","email":"","orcid":"","institution":"Sorbonne Université, Institut du Cerveau – Paris Brain Institute – ICM, INSERM 1127, CNRS 7225, APHP-Hôpital de la Pitié Salpêtrière, MOV’IT Team","correspondingAuthor":false,"prefix":"","firstName":"Cécile","middleName":"","lastName":"Gallea","suffix":""},{"id":518836276,"identity":"ac6cfb0e-db73-4233-a8a6-f55a15f5dc6b","order_by":5,"name":"Martina Bracco","email":"","orcid":"","institution":"Sorbonne Université, Institut du Cerveau – Paris Brain Institute – ICM, INSERM 1127, CNRS 7225, APHP-Hôpital de la Pitié Salpêtrière, MOV’IT Team","correspondingAuthor":false,"prefix":"","firstName":"Martina","middleName":"","lastName":"Bracco","suffix":""},{"id":518836277,"identity":"3ba5cc37-dab9-40eb-b8e6-e9572021e2e9","order_by":6,"name":"Antoni Valero-Cabré","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABU0lEQVRIie2RMUvDQBSAXzm4LE+yHliSv3DlwCqI/pWGQCcthYIIFlooxMXS9cQ/0anzhYN0qXbN4FDJ4uCQ4lQo6NlWU21wFswHB++O9917jwdQUPBn4eZY67AGQJQyAX7cVG462SjkS6G1jULAyDxfgW8K8uwpR7HvBmHSbII36JEoTdsnjer1zVwtzqBctaehngdLB6x+8pwp7FETITl4UlP/VkZ+qzy5H4b9EeCR9EGFAReA44PDrTKxT/dNJwI0CrLXJZ5k50NdGr0hj8lK8bqsTrfacz8VV9uvRul40n2ZGcVMNNUrpfND4RvF4RqJUbSpgrBW1LqxGrA6mWVKJfaFQM6ciqaiJKNxi2Gdr2Yxv3E1eRCVACO6NYoTe08JLo/RGfcSSNtXDWbpJF2M4JRPwyS9vHBc2wpIurMctruvDFOC/pqQS06VgoKCgv/DO6v+ebOa1BhuAAAAAElFTkSuQmCC","orcid":"","institution":"Boston University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Antoni","middleName":"","lastName":"Valero-Cabré","suffix":""}],"badges":[],"createdAt":"2025-09-11 15:38:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7593453/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7593453/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92152647,"identity":"a7b684ce-422a-4d26-9e46-829c469f6c02","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2860996,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptefieldspatialfilteringCorominasetal.LirTMS2025V.9.docx","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/9134d96df7bee358b6275b58.docx"},{"id":92152638,"identity":"dcc9b4db-d28d-4d90-82fd-e9ddd0d7ac4a","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8791,"visible":true,"origin":"","legend":"","description":"","filename":"f452a2931b5b43dd9949509aa38dedf3.json","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/609b4988e04e8f6dd3fbbf23.json"},{"id":92152643,"identity":"1b14ddd8-a2a8-4a9a-9ebf-da8349e61ae0","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":70837,"visible":true,"origin":"","legend":"","description":"","filename":"f452a2931b5b43dd9949509aa38dedf31enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/ef3f4948856c0246bed98800.xml"},{"id":92152642,"identity":"f3af5266-3516-4ec3-89ef-a56f8d9d06c7","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":439993,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/b29f3063a3d2497bb5077272.png"},{"id":92152641,"identity":"0e5e4151-6073-418b-ac87-c01b25def0dd","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":539637,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/c34001c1c4766500251e2d00.png"},{"id":92152967,"identity":"a30fac32-4cd9-4a09-8911-8009ff92fda9","added_by":"auto","created_at":"2025-09-25 08:36:58","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76471,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/499f0c3f3dcf7821b2563622.png"},{"id":92152645,"identity":"2413935b-e667-4385-974a-5c13fe7c9817","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138438,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/232bdc6ebc69c2ed921c8f96.png"},{"id":92152968,"identity":"39241c10-c1de-4177-a9b9-e6ffda05043e","added_by":"auto","created_at":"2025-09-25 08:36:58","extension":"xml","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69340,"visible":true,"origin":"","legend":"","description":"","filename":"f452a2931b5b43dd9949509aa38dedf31structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/a795ac85a63e389445b8624c.xml"},{"id":92152646,"identity":"d2a78896-b8b0-42ff-aaa3-c624dcea513a","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"html","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":77288,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/fff7f1643e49b72ec4bfd9a4.html"},{"id":92152637,"identity":"6069775c-9b5f-4f30-be54-7800eb7f8140","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":150548,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMethodological overview\u003c/strong\u003e. Illustrative step-by-step procedure of the E-field-based spatial filtering pipeline EPICURUS. (\u003cstrong\u003ea\u003c/strong\u003e) Raw TMS-EEG data (\u003cstrong\u003eb\u003c/strong\u003e) were preprocessed and cleaned to remove artifacts from each of the n=18 participant datasets. (\u003cstrong\u003ec\u003c/strong\u003e) In parallel, MRI-based head models and E-field distributions on each MRI volume were computed for each participant. (\u003cstrong\u003ed\u003c/strong\u003e) Finally, ROIs corresponding to the stimulated left M1 and derived from \u0026gt;80% E-Field strength model were used to reconstruct the signal directly evoked by TMS. (\u003cstrong\u003ee\u003c/strong\u003e) From the preprocessed data, a minimum norm-based (MNE) reconstruction was first estimated. This was followed by the implementation of an estimator with noise and distributed source constraints to minimize crosstalk leakage from any sources other than those from our ROI. CTF: CrossTalk Function employed for spatial filtering; FDI: First Dorsal Interosseous; MNE: Minimum Norm Estimator; TMS-EEG: concurrent Transcranial Magnetic Stimulation with ElectroEncephaloGraphy.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/744fbf8a12740175d6f6bbf2.png"},{"id":92152636,"identity":"5faad4dc-b7ce-4bc8-851f-5419b92a74ee","added_by":"auto","created_at":"2025-09-25 08:28:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":301467,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGroup average results\u003c/strong\u003e. \u003cstrong\u003e(a)\u003c/strong\u003e \u003cstrong\u003eReference data\u003c/strong\u003e. \u003cstrong\u003e1\u003c/strong\u003e: Sensor-level time-course across the 64-channels corresponding to “reference data” TEPs. \u003cstrong\u003e2:\u003c/strong\u003e Sensor-level Butterfly plot (superior panel) and topographic sensor and source-level projection (inferior panel) across the 64-channels corresponding to the original “reference data” TEPs. \u003cstrong\u003e3\u003c/strong\u003e: Sensor-level original “reference data” TEPs from a cluster of electrodes under the stimulated left M1 (mean of C1, C3,CP1,CP3; sensor-level original; blue trace), and from the region most impacted by TMS (ROI) using a standard MNE (mean signal from the ROI; source-level original; red trace). Group-level source space data are visualized by projecting group-averaged signals into a standard MNI152 mesh.\u003cstrong\u003e (b)\u003c/strong\u003e \u003cstrong\u003eFilters’ validation\u003c/strong\u003e. Statistical comparisons of original “reference data” sensor- and source-level (blue and red traces respectively) and E-field-based filtered TEPs (yellow trace) (left panel). Statistical comparisons of original “reference data” sensor- and source-level (blue and red traces, respectively) and E-field-based filtered TEPs combined with auditory constraints (black trace; central panel). Statistical comparisons of filtered (with and without constraints; green and yellow traces, respectively) TEPs (right panel). TEPs group mean and standard deviation across subjects are shown as a thick centered line and a background shaded line, respectively. Vertical grey bars highlight statistically significant clusters (p\u0026lt;0.05).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/e546e6d8d765d18dca568785.png"},{"id":95654979,"identity":"dea4a2cf-25fb-4156-939a-c472eecbe5d4","added_by":"auto","created_at":"2025-11-11 16:13:56","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":909417,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7593453/v1/d8d9df09-c621-4b94-a1a3-d721771868c0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"EPICURUS: E-field-based spatial filtering procedure for an accurate estimation of local EEG activity evoked by Transcranial Magnetic Stimulation","fulltext":[{"header":"HIGHLIGHTS","content":"\u003cul\u003e\n \u003cli\u003eThe lack of spatial specificity of EEG prevents an accurate estimation of the locally activated EEG sources, hindering the interpretation of TMS effects.\u003c/li\u003e\n \u003cli\u003eWe here developed a spatial filtering approach for TMS-EEG based on individual estimation of E-field spread following stimulation. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eOur E-field spatial filter, combined with auditory constraints, reduced non-relevant distant signal components and improved the reliability of local EEG estimates.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"INTRODUCTION","content":"\u003cp\u003eElectroencephalography (EEG) is a neurophysiological approach increasingly employed to monitor the impact of Transcranial Magnetic Stimulation (TMS) on brain activity. However, the activity recorded by such a high-temporal resolution technique after a TMS pulse remains spatially unspecific and reflects a combination of mixed cerebral sources directly or indirectly activated by TMS. This phenomenon occurs because, besides relevant local direct effects on the targeted cortical site, TMS engages activity in interconnected distant areas and networks [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], while also activating non-relevant regions due to TMS-generated sensory input in the auditory, tactile, and vestibular domains [\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eHere, we present an innovative approach to isolate activity locally elicited by TMS pulses within the targeted cortical region, defined by the modeled electric field extension, while minimizing unspecific contributions from non-relevant (directly and indirectly) activated sources. The method (EPICURUS; \u003cb\u003eE\u003c/b\u003e-field based s\u003cb\u003eP\u003c/b\u003eatial f\u003cb\u003eI\u003c/b\u003eltering for \u003cb\u003eC\u003c/b\u003eoncurrent EEG-TMS data \u003cb\u003eU\u003c/b\u003esed to ta\u003cb\u003eR\u003c/b\u003eget so\u003cb\u003eU\u003c/b\u003erce \u003cb\u003eS\u003c/b\u003eignal) has been designed as a spatial filter for TMS-EEG datasets and combines E-field simulations [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] with a Minimum Norm-like EEG source Estimator (MNE), minimizing signal leakage and crosstalk from distal brain locations [\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eParticipants\u003c/h2\u003e\u003cp\u003eTo test EPICURUS performance, we used it to analyze a TMS-EEG dataset of n\u0026thinsp;=\u0026thinsp;18 healthy participants (10 females, 8 males, 26\u0026thinsp;\u0026plusmn;\u0026thinsp;5 years old), all right-handed, meeting safety criteria for MRI [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] and TMS [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], who provided written consent to participate in the study. Our protocol followed the Declaration of Helsinki and received approval from a local ethical committee (\u003cem\u003eComit\u0026eacute; de Protection des Personnes, Ile de France I\u003c/em\u003e).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eTMS-EEG protocol and data processing\u003c/h3\u003e\n\u003cp\u003eParticipants received 80 single TMS pulses (with a variable inter-stimulus interval of 5-7sec) over the left primary motor cortex (M1) while we continuously recorded scalp EEG (25kHz sampling rate; ActiChamp, with 65 electrodes placed following the International 10/10 system coordinates). The left M1 'hotspot\u0026rsquo; was defined as the optimal cortical location for eliciting Motor-Evoked Potentials in the First Dorsal Interosseous muscle at rest [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. TMS was administered with a D70 mm figure-of-eight alpha coil attached to a biphasic stimulator (Magstim Rapid2) at a fixed intensity (60% of the maximal stimulator output; MSO). For each participant, the TMS coil position was maintained consistently throughout the experiment using an MRI-frameless stereotactic system (Brainsight, Rogue Solutions Inc., Montr\u0026eacute;al, QC, Canada).\u003c/p\u003e\u003cp\u003eThe TMS-EEG dataset was processed using the MATLAB-based toolbox EEGLAB [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] and its TMS\u0026ndash;EEG signal analyzer plugin (TESA [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]). Data was epoched around the TMS pulse (from \u0026minus;\u0026thinsp;1200 to 1200ms \u0026ndash; w.r.t. \u0026ndash; the TMS pulse) and baseline corrected (from \u0026minus;\u0026thinsp;500 to -10ms). TMS-pulse electromagnetic (from \u0026minus;\u0026thinsp;2ms to 12ms w.r.t the TMS pulse) and recharging artifacts (visually identified [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]) were removed and interpolated using a spline function. The data were downsampled to 1 kHz, and the most artifacted channels were temporarily removed from analyses, and the independent component analysis (ICA) was employed to remove ocular artifacts. Next, the SOUND algorithm was used for noise suppression, replacement of removed channels, and signal re-referencing to a common average [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The SSP-SIR algorithm was then used to suppress TMS-evoked muscle artifacts [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], and a second ICA was computed to remove residual continuous muscle artifacts. Finally, the data were bandpass filtered with a fourth-order Butterworth filter (2-45Hz) and baseline corrected.\u003c/p\u003e\n\u003ch3\u003eE-field-based spatial filter\u003c/h3\u003e\n\u003cp\u003eOur pipeline integrates a previously developed spatial filter approach based on cross-talk functions (DeFleCT; [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]) with E-field distribution modeled using SimNIBS 4.0 [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. First, we reconstructed individual finite element head models (FEM [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]) for each participant and computed the leadfield matrix employing the Helsinki BEM framework from the FEM model [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. We simulated the E-field distribution induced by TMS at the intensity of 60% MSO with SimNIBS (version 4.0, 68,8dI/dtmax\u0026thinsp;=\u0026thinsp;60% MSO; [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]), using TMS coil positions recorded during the experimental session. We employed each participant\u0026rsquo;s E-field spatial distribution to identify the cortical area most impacted by TMS (\u0026gt;\u0026thinsp;80% E-field peak strength; region of interest\u0026ndash;ROI). Next, a spatial filter with distributed source minimization constraints (i.e., crosstalk from outside the established ROI) was built for each participant. The spatial filtering strategy mirrors Special Case 2 of the DeFleCT implementation [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], using an L2 minimum-norm estimate with a normalization constraint tailored to the TMS targeted sources combined with a noise minimization constraint (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). To validate our approach and explore the efficacy of the filter, we implemented an additional filtering procedure; a discrete source constraint minimizing signal components arising from sources in the auditory cortex (~\u0026thinsp;Brodman areas 22,41, and 42; [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]; bilaterally delineated on each individual head model). Hence, the original EEG signals were filtered to generate two dataset conditions: one using the E-field-based filter alone and another adding to the former the above-mentioned auditory constraints.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eSpatial filter validation and statistical analysis\u003c/h3\u003e\n\u003cp\u003eTo validate our spatial filtering pipeline, we first identified TEPs from all available EEG sensors (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.A.1, A.2) and characterized their latency, morphology, and topographies. Next, we analyzed TEPs from a cluster of electrodes in the vicinity of the TMS stimulation site (i.e., mean signal from C1, C3, CP1, and CP3; sensor-space original signal; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.A.3). Additionally, we employed a classical MNE estimator, without noise minimization (i.e., source-level original signal), to reconstruct EEG time-series directly from the sources impacted by the stimulation (i.e., the sources identified within the E-field-simulation-based ROI; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.A.3).\u003c/p\u003e\u003cp\u003eThe sensor-level and source-level original TEPs served as \u003cem\u003ereference data\u003c/em\u003e conditions for our analysis and were statistically compared to the spatially filtered signals (E-field-based and E-field-based with auditory constraints). Comparisons were performed with non-parametric cluster-based permutation statistics employing one-way 1x3 ANOVA (FieldTrip; [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]; 50000 iterations, α\u0026thinsp;=\u0026thinsp;0.05, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, time-window: from 0ms to 400ms -w.r.t. TMS pulse). Whenever needed, post-hoc tests were conducted using cluster-based non-parametric statistics with two-tailed dependent-samples t-tests (50000 iterations, α\u0026thinsp;=\u0026thinsp;0.05, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), employing Bonferroni correction for multiple comparisons.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eComparisons between E-field-based filtered datasets and the original source-level and sensor-level signals revealed significant differences, identifying a significant cluster\u0026thinsp;~\u0026thinsp;175ms following TMS pulse (\u003cem\u003eF\u003c/em\u003e(2, 32)\u0026thinsp;=\u0026thinsp;9.1, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.B, left panel). Post-hoc comparisons indicated significant differences between the source-level and the sensor-level original signals (paired t-test: \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.6, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.006), and between the E-field-based filtered signal and sensor-level original signal (paired t-test: \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.9, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.04). However, no significant differences were observed between the E-field-based filtered signal and the original source-level dataset (paired t-test: \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.3, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.3).\u003c/p\u003e\u003cp\u003eComparisons between E-field-based filtered signals with auditory constraints, the original source-level and sensor-level signals revealed significant differences across signals and revealed two clusters with different latencies (cluster 1, ~\u0026thinsp;115ms, \u003cem\u003eF\u003c/em\u003e(2, 32)\u0026thinsp;=\u0026thinsp;7.7, \u003cem\u003ep\u0026thinsp;=\u003c/em\u003e\u0026thinsp;0.03; cluster 2, ~\u0026thinsp;175ms, \u003cem\u003eF\u003c/em\u003e(2, 32)\u0026thinsp;=\u0026thinsp;9.7, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.B, central panel). Post-hoc comparisons indicated significant differences between the E-field-based filtered with auditory constraints and the sensor level original signal on both clusters (cluster 1, paired t-test: ~115ms, \u003cem\u003et\u003c/em\u003e = -4, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01 / cluster 2, paired t-test: ~175ms, \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.2, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01), and between the E-field-based filtered with auditory constraints and the source level original signal (paired t-test: ~115ms, \u003cem\u003et\u003c/em\u003e = -2.7, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.05).\u003c/p\u003e\u003cp\u003eFinally, the comparison between the E-field-based filtered signal and the E-field-based with auditory constraints conditions revealed significant differences in the same two late latency clusters described in the previous analysis (cluster 1, paired t-test: ~115ms, \u003cem\u003et\u003c/em\u003e = -4.5, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0004; cluster 2, paired t-test: ~175ms, \u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.6, \u003cem\u003ep corrected\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.B, right panel).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eWe designed and implemented EPICURUS, a spatial filtering approach for TMS-EEG data that employs the TMS coil position during the experimental session, the E-field distribution biophysical model, and an MNE source reconstruction method minimizing signal leakage and crosstalk from distant brain sources.\u003c/p\u003e\u003cp\u003eOur results revealed three key findings. First, the shape and latency of E-field-based filtered TEPs (with and without auditory constraints) were comparable to those originally recorded or estimated (source and sensor level), supporting the reliability and robustness of an approach that does not alter the overall time course of TMS-evoked signals. Second, E-field-based filtered TEPs (with and without auditory constraints) showed lower amplitude than the sensor-level original TEPs, especially at later components (\u0026gt;\u0026thinsp;100 ms post TMS-pulse). However, only TEPs processed with EPICURUS including auditory constraints reached statistically significant differences compared to original source-level TEPs. Third, signals reconstructed with the E-field-based filtering method differed from those reconstructed with E-field-based filtering combined with auditory constraints, particularly\u0026thinsp;~\u0026thinsp;115ms.\u003c/p\u003e\u003cp\u003ePrior work has shown that late TEP components, such as those present in our data at ~\u0026thinsp;115 and ~\u0026thinsp;175 ms (N115 and P175) with fronto-central topographies (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.A.2), typically reflect the engagement of distant brain regions indirectly activated by TMS (i.e., multisensory responses [\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]). We here hypothesize that statistical differences between filtered and original signals reported may arise from the attenuation of distant sources, which, at least in part, reflect the indirect multisensory EEG responses to TMS. A likely direct link to auditory evoked activity is supported by our findings showing that the addition of auditory constraints significantly reduces the amplitudes of such late TEP components.\u003c/p\u003e\u003cp\u003eAlthough our filtering approach efficiently suppressed the contribution of TMS activity associated with multisensory responses, the primary aim of EPICURUS is to increase the reliability of local evoked responses by minimizing the distant non-relevant sources. The suppression of multisensory responses in our TMS-EEG example is an indirect consequence of this aim. To effectively minimize multisensory responses, proper experimental procedures during experimental sessions (e.g., noise masking, control conditions), potentially combined with other analysis strategies, will be necessary [\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cu\u003eCOMPETING INTERESTS\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing financial interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eDATA AND CODE AVAILABILITY\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets produced and/or analyzed during the current study cannot be made publicly available due to institutional regulations. However, they can be obtained from the corresponding and/or first authors of the manuscript upon reasonable request. MATLAB codes for the pre-processing and filtering implementation can be found at the permanent link:\u0026nbsp;\u003cem\u003ehttps://github.com/XavierCorominas/Source-Based-filters-for-brain-stimulation-readouts\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eCREDIT AUTHORSHIP CONTRIBUTION STATEMENT\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: XC-T, TM, MB, AV-C. Methodology: XC-T, TM, MB. Data curation and analysis: XC-T. Funding acquisition: CG, MB, AV-C. Supervision: MB, AV-C. Writing \u0026ndash; original draft: XC-T, AV-C \u0026amp; MB. Writing \u0026ndash; review and editing: XC-T, TM, CL, CG, MT-C, MB, AV-C.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cu\u003eACKNOWLEDGEMENTS\u003c/u\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by\u0026nbsp;the Agence Nationale de la Recherche in France (BrainMAG; ANR-19-CE37-0021; AV-C), the European Union\u0026rsquo;s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement 897941 (MB), the Big Brain Theory (BBT-3; FORTE) internal ICM grant (CG, AV-C \u0026amp; MB), ICM CARNOT training grant (MB, CG), IHU-ICM CARNOT Maturation grant, and additional funding came from Investissements d\u0026rsquo;avenir (ANR-10- IAIHU-0006) awarded to XC-T and AV-C for associated projects.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBortoletto M, Veniero D, Thut G, Miniussi C (2015) The contribution of TMS\u0026ndash;EEG coregistration in the exploration of the human cortical connectome. Neurosci Biobehav Rev 49:114\u0026ndash;124. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neubiorev.2014.12.014\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2014.12.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBracco M, Mutanen TP, Veniero D, Thut G, Robertson EM (2025) Protocol to assess changes in brain network resistance to perturbation during offline processing using TMS-EEG. STAR Protoc 6:103622. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.xpro.2025.103622\u003c/span\u003e\u003cspan address=\"10.1016/j.xpro.2025.103622\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRocchi L, Di Santo A, Brown K, Ib\u0026aacute;\u0026ntilde;ez J, Casula E, Rawji V et al (2021) Disentangling EEG responses to TMS due to cortical and peripheral activations. Brain Stimulation: Basic Translational Clin Res Neuromodulation 14:4\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.brs.2020.10.011\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2020.10.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eConde V, Tomasevic L, Akopian I, Stanek K, Saturnino GB, Thielscher A et al (2019) The non-transcranial TMS-evoked potential is an inherent source of ambiguity in TMS-EEG studies. NeuroImage 185:300\u0026ndash;312. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neuroimage.2018.10.052\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2018.10.052\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGordon PC, Desideri D, Belardinelli P, Zrenner C, Ziemann U (2018) Comparison of cortical EEG responses to realistic sham versus real TMS of human motor cortex. Brain Stimulation: Basic Translational Clin Res Neuromodulation 11:1322\u0026ndash;1330. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.brs.2018.08.003\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2018.08.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHernandez-Pavon JC, Veniero D, Bergmann TO, Belardinelli P, Bortoletto M, Casarotto S et al (2023) TMS combined with EEG: Recommendations and open issues for data collection and analysis. Brain Stimul 16:567\u0026ndash;593. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.brs.2023.02.009\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2023.02.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiessen E, Bracco M, Mutanen TP, Robertson EM (2021) An analytical approach to identify indirect multisensory cortical activations elicited by TMS? Brain Stimulation: Basic, Translational, and Clinical Research in Neuromodulation. 14(2):376\u0026ndash;378. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.brs.2021.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2021.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eThielscher A, Antunes A, Saturnino GB (2015) Field modeling for transcranial magnetic stimulation: A useful tool to understand the physiological effects of TMS? 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 222\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/EMBC.2015.7318340\u003c/span\u003e\u003cspan address=\"10.1109/EMBC.2015.7318340\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHauk O, Stenroos M (2014) A framework for the design of flexible cross-talk functions for spatial filtering of EEG/MEG data: DeFleCT. Hum Brain Mapp 35:1642\u0026ndash;1653. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1002/hbm.22279\u003c/span\u003e\u003cspan address=\"10.1002/hbm.22279\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStenroos M, Hunold A, Haueisen J (2014) Comparison of three-shell and simplified volume conductor models in magnetoencephalography. NeuroImage 94:337\u0026ndash;348. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/\u003c/span\u003e\u003cspan address=\"https://doi.org/https://doi.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuroimage.2014.01.006\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2014.01.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eH\u0026auml;m\u0026auml;l\u0026auml;inen MS, Ilmoniemi RJ (1994) Interpreting magnetic fields of the brain: minimum norm estimates. Med Biol Eng Comput 32:35\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF02512476\u003c/span\u003e\u003cspan address=\"10.1007/BF02512476\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShellock FG, Spinazzi A, MRI Safety Update (2008) : Part 2, Screening Patients for MRI. American Journal of Roentgenology 2008;191:1140\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2214/AJR.08.1038.2\u003c/span\u003e\u003cspan address=\"10.2214/AJR.08.1038.2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRossi S, Hallett M, Rossini PM, Pascual-Leone A (2009) Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clin Neurophysiol 120:2008\u0026ndash;2039. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.clinph.2009.08.016\u003c/span\u003e\u003cspan address=\"10.1016/j.clinph.2009.08.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen R, Cros D, Curra A, Di Lazzaro V, Lefaucheur J-P, Magistris MR et al (2008) The clinical diagnostic utility of transcranial magnetic stimulation: Report of an IFCN committee. Clin Neurophysiol 119:504\u0026ndash;532. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.clinph.2007.10.014\u003c/span\u003e\u003cspan address=\"10.1016/j.clinph.2007.10.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDelorme A, Makeig S (2004) EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods 134:9\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.jneumeth.2003.10.009\u003c/span\u003e\u003cspan address=\"10.1016/j.jneumeth.2003.10.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRogasch NC, Sullivan C, Thomson RH, Rose NS, Bailey NW, Fitzgerald PB et al (2017) Analysing concurrent transcranial magnetic stimulation and electroencephalographic data: A review and introduction to the open-source TESA software. NeuroImage 147:934\u0026ndash;951. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neuroimage.2016.10.031\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2016.10.031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVeniero D, Bortoletto M, Miniussi C (2009) TMS-EEG co-registration: On TMS-induced artifact. Clin Neurophysiol 120:1392\u0026ndash;1399. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/\u003c/span\u003e\u003cspan address=\"https://doi.org/https://doi.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.clinph.2009.04.023\u003c/span\u003e\u003cspan address=\"10.1016/j.clinph.2009.04.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMutanen TP, Metsomaa J, Liljander S, Ilmoniemi RJ (2018) Automatic and robust noise suppression in EEG and MEG: The SOUND algorithm. NeuroImage 166:135\u0026ndash;151. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neuroimage.2017.10.021\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2017.10.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMutanen TP, Kukkonen M, Nieminen JO, Stenroos M, Sarvas J, Ilmoniemi RJ (2016) Recovering TMS-evoked EEG responses masked by muscle artifacts. NeuroImage 139:157\u0026ndash;166. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neuroimage.2016.05.028\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2016.05.028\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePuonti O, Van Leemput K, Saturnino GB, Siebner HR, Madsen KH, Thielscher A (2020) Accurate and robustwhole-head segmentation from magnetic resonance images for individualized head modeling. NeuroImage 219:117044. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neuroimage.2020.117044\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2020.117044\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDrakaki M, Mathiesen C, Siebner HR, Madsen K, Thielscher A (2022) Database of 25 validated coil models for electric field simulations for TMS. Brain Stimul 15:697\u0026ndash;706. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.brs.2022.04.017\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2022.04.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHackett TA (2015) Chapter 2 - Anatomic organization of the auditory cortex. In: Aminoff MJ, Boller F, Swaab DF, editors. Handb Clin Neurol, vol. 129, Elsevier; pp. 27\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/B978-0-444-62630-1.00002-0\u003c/span\u003e\u003cspan address=\"10.1016/B978-0-444-62630-1.00002-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOostenveld R, Fries P, Maris E, Schoffelen J-M, FieldTrip (2011) Open Source Software for Advanced Analysis of MEG, EEG, and Invasive Electrophysiological Data. Comput Intell Neurosci 2011:156869. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1155/2011/156869\u003c/span\u003e\u003cspan address=\"10.1155/2011/156869\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaris E, Oostenveld R (2007) Nonparametric statistical testing of EEG- and MEG-data. J Neurosci Methods 164:177\u0026ndash;190. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/\u003c/span\u003e\u003cspan address=\"https://doi.org/https://doi.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jneumeth.2007.03.024\u003c/span\u003e\u003cspan address=\"10.1016/j.jneumeth.2007.03.024\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMetsomaa J, Song Y, Mutanen TP, Gordon PC, Ziemann U, Zrenner C et al Adapted Beamforming: A Robust and Flexible Approach for Removing Various Types of Artifacts from TMS\u0026ndash;EEG Data. Brain Topogr 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10548-024-01044-4\u003c/span\u003e\u003cspan address=\"10.1007/s10548-024-01044-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHernandez-Pavon JC, Kugiumtzis D, Zrenner C, Kimiskidis VK, Metsomaa J (2022) Removing artifacts from TMS-evoked EEG: A methods review and a unifying theoretical framework. J Neurosci Methods 376:109591. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.jneumeth.2022.109591\u003c/span\u003e\u003cspan address=\"10.1016/j.jneumeth.2022.109591\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBiabani M, Fornito A, Mutanen TP, Morrow J, Rogasch NC (2019) Characterizing and minimizing the contribution of sensory inputs to TMS-evoked potentials. Brain Stimul 12:1537\u0026ndash;1552. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.brs.2019.07.009\u003c/span\u003e\u003cspan address=\"10.1016/j.brs.2019.07.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Brain Stimulation, Electrical field simulation, Transcranial magnetic stimulation, concurrent TMS-EEG, spatial filtering","lastPublishedDoi":"10.21203/rs.3.rs-7593453/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7593453/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: The concurrent use of Transcranial magnetic stimulation (TMS) with electroencephalography (EEG) is increasingly integrated into research and clinical protocols. However, a reliable isolation of EEG responses that are locally evoked by TMS at the targeted cortical sites remains challenging.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: We introduce EPICURUS, a novel spatial filtering approach for TMS-EEG that uses individualized MRI-based simulations of the TMS-induced electric field (E-field) to define the spatial extent of locally evoked activity. This method guides the reconstruction of EEG signals originating from the primary stimulation site while minimizing crosstalk from distant, non-targeted sources.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Application of EPICURUS revealed a marked reduction in late transcranial evoked potential (TEP) components (\u0026gt;100 ms), indicating that E-field-based spatial filtering effectively suppresses contributions from indirect or multisensory sources, particularly when combined with auditory suppression techniques.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: By leveraging the spatial precision of individualized E-field modeling, EPICURUS enhances the specificity of EEG signal reconstruction, offering a promising tool for improving the spatiotemporal resolution of cortical responses directly evoked by TMS.\u003c/p\u003e","manuscriptTitle":"EPICURUS: E-field-based spatial filtering procedure for an accurate estimation of local EEG activity evoked by Transcranial Magnetic Stimulation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-25 08:28:53","doi":"10.21203/rs.3.rs-7593453/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"64bd9049-b894-4ead-84d2-2aedb4861540","owner":[],"postedDate":"September 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-20T15:10:16+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-25 08:28:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7593453","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7593453","identity":"rs-7593453","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}