Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach
Brain state-dependent transcranial magnetic stimulation (TMS) holds promise for enhancing neuromodulatory effects by synchronizing stimulation with specific features of cortical oscillations derived from real-time electroencephalography (EEG). However, conventional approaches rely on open-loop syste...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10795227/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850242969957826560 |
|---|---|
| author | Lisa Haxel Oskari Ahola Paolo Belardinelli Maria Ermolova Dania Humaidan Jakob H. Macke Ulf Ziemann |
| author_facet | Lisa Haxel Oskari Ahola Paolo Belardinelli Maria Ermolova Dania Humaidan Jakob H. Macke Ulf Ziemann |
| author_sort | Lisa Haxel |
| collection | DOAJ |
| description | Brain state-dependent transcranial magnetic stimulation (TMS) holds promise for enhancing neuromodulatory effects by synchronizing stimulation with specific features of cortical oscillations derived from real-time electroencephalography (EEG). However, conventional approaches rely on open-loop systems with static stimulation parameters, assuming that pre-determined EEG features universally indicate high or low excitability states. This one-size-fits-all approach overlooks individual neurophysiological differences and the dynamic nature of brain states, potentially compromising therapeutic efficacy. We present a supervised machine learning framework that predicts individual motor excitability states from pre-stimulus EEG features. Our approach combines established biomarkers with a comprehensive set of spectral and connectivity measures, implementing multi-scale feature selection within a nested cross-validation scheme. Validation across multiple classifiers, feature sets, and experimental protocols in 50 healthy participants demonstrated a mean prediction accuracy of <inline-formula> <tex-math notation="LaTeX">$71 \; \pm \; 7$ </tex-math></inline-formula>%. Hierarchical clustering of top predictive EEG features revealed two distinct participant subgroups. The first subgroup, comprising approximately 50% of participants, showed predictive features predominantly in alpha and low-beta bands in sensorimotor regions of the stimulated hemisphere, aligning with traditional associations of motor excitability and the sensorimotor <inline-formula> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula>-rhythm. The second subgroup exhibited predictive features primarily in low and high gamma bands in parietal regions, suggesting that motor excitability is influenced by broader neural dynamics for these individuals. Our data-driven framework effectively identifies personalized motor excitability biomarkers, holding promise to optimize TMS interventions in clinical and research settings. Additionally, our approach provides a versatile platform for biomarker discovery and validation across diverse neuromodulation paradigms and brain signal classification tasks. |
| format | Article |
| id | doaj-art-2e1b9fa8f2e3401faa11619ac326ff98 |
| institution | OA Journals |
| issn | 1534-4320 1558-0210 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj-art-2e1b9fa8f2e3401faa11619ac326ff982025-08-20T02:00:08ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1534-43201558-02102025-01-013310311210.1109/TNSRE.2024.351639310795227Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning ApproachLisa Haxel0https://orcid.org/0009-0003-3897-3713Oskari Ahola1Paolo Belardinelli2Maria Ermolova3Dania Humaidan4https://orcid.org/0000-0003-1381-257XJakob H. Macke5Ulf Ziemann6https://orcid.org/0000-0001-8372-3615Department of Neurology and Stroke, the Excellence Cluster Machine Learning, and Tübingen AI Center, University of Tübingen, Tübingen, GermanyHertie Institute for Clinical Brain Research, Tübingen, GermanyDepartment of Neurology and Stroke, University of Tübingen, Tübingen, GermanyHertie Institute for Clinical Brain Research, Tübingen, GermanyHertie Institute for Clinical Brain Research, Tübingen, GermanyExcellence Cluster Machine Learning and Tübingen AI Center, University of Tübingen, Tübingen, GermanyHertie Institute for Clinical Brain Research, Tübingen, GermanyBrain state-dependent transcranial magnetic stimulation (TMS) holds promise for enhancing neuromodulatory effects by synchronizing stimulation with specific features of cortical oscillations derived from real-time electroencephalography (EEG). However, conventional approaches rely on open-loop systems with static stimulation parameters, assuming that pre-determined EEG features universally indicate high or low excitability states. This one-size-fits-all approach overlooks individual neurophysiological differences and the dynamic nature of brain states, potentially compromising therapeutic efficacy. We present a supervised machine learning framework that predicts individual motor excitability states from pre-stimulus EEG features. Our approach combines established biomarkers with a comprehensive set of spectral and connectivity measures, implementing multi-scale feature selection within a nested cross-validation scheme. Validation across multiple classifiers, feature sets, and experimental protocols in 50 healthy participants demonstrated a mean prediction accuracy of <inline-formula> <tex-math notation="LaTeX">$71 \; \pm \; 7$ </tex-math></inline-formula>%. Hierarchical clustering of top predictive EEG features revealed two distinct participant subgroups. The first subgroup, comprising approximately 50% of participants, showed predictive features predominantly in alpha and low-beta bands in sensorimotor regions of the stimulated hemisphere, aligning with traditional associations of motor excitability and the sensorimotor <inline-formula> <tex-math notation="LaTeX">$\mu $ </tex-math></inline-formula>-rhythm. The second subgroup exhibited predictive features primarily in low and high gamma bands in parietal regions, suggesting that motor excitability is influenced by broader neural dynamics for these individuals. Our data-driven framework effectively identifies personalized motor excitability biomarkers, holding promise to optimize TMS interventions in clinical and research settings. Additionally, our approach provides a versatile platform for biomarker discovery and validation across diverse neuromodulation paradigms and brain signal classification tasks.https://ieeexplore.ieee.org/document/10795227/Brain state classificationelectroencephalography (EEG)machine learningmotor excitabilitytranscranial magnetic stimulation (TMS) |
| spellingShingle | Lisa Haxel Oskari Ahola Paolo Belardinelli Maria Ermolova Dania Humaidan Jakob H. Macke Ulf Ziemann Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach IEEE Transactions on Neural Systems and Rehabilitation Engineering Brain state classification electroencephalography (EEG) machine learning motor excitability transcranial magnetic stimulation (TMS) |
| title | Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach |
| title_full | Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach |
| title_fullStr | Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach |
| title_full_unstemmed | Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach |
| title_short | Decoding Motor Excitability in TMS Using EEG-Features: An Exploratory Machine Learning Approach |
| title_sort | decoding motor excitability in tms using eeg features an exploratory machine learning approach |
| topic | Brain state classification electroencephalography (EEG) machine learning motor excitability transcranial magnetic stimulation (TMS) |
| url | https://ieeexplore.ieee.org/document/10795227/ |
| work_keys_str_mv | AT lisahaxel decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT oskariahola decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT paolobelardinelli decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT mariaermolova decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT daniahumaidan decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT jakobhmacke decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach AT ulfziemann decodingmotorexcitabilityintmsusingeegfeaturesanexploratorymachinelearningapproach |