Improving TMS-EEG Signal Quality for Closed-Loop Neurostimulation via Source-Domain Denoising
Researchers have developed and validated a new preprocessing pipeline designed to enhance the quality of Transcranial Magnetic Stimulation combined with Electroencephalography (TMS-EEG) signals. The study addresses the significant challenge of artifact removal in TMS-EEG data, which is critical for accurate analysis in both clinical and research settings. By evaluating two widely used source-based artifact removal approaches, the team established a reference benchmark dataset to support future algorithm development and enable systematic comparisons, despite the lack of a true physiological ground truth. The results demonstrate that the proposed workflow robustly improves signal quality while preserving TMS-evoked potentials. This advancement is pivotal for integrating TMS-EEG into larger Brain-Computer Interface (BCI) frameworks, ultimately aiming to deepen the understanding of cortical dynamics. The findings suggest improved data reliability for closed-loop neurostimulation applications, potentially expanding the utility of TMS-EEG in diagnosing and treating neurological conditions through more precise monitoring and intervention strategies.
Wire timeline
Improving TMS-EEG Signal Quality for Closed-Loop Neurostimulation via Source-Domain Denoising
Researchers have developed and validated a new preprocessing pipeline designed to enhance the quality of Transcranial Magnetic Stimulation combined with Electroencephalography (TMS-EEG) signals. The study addresses the significant challenge of artifact removal in TMS-EEG data, which is critical for accurate analysis in both clinical and research settings. By evaluating two widely used source-based artifact removal approaches, the team established a reference benchmark dataset to support future algorithm development and enable systematic comparisons, despite the lack of a true physiological ground truth. The results demonstrate that the proposed workflow robustly improves signal quality while preserving TMS-evoked potentials. This advancement is pivotal for integrating TMS-EEG into larger Brain-Computer Interface (BCI) frameworks, ultimately aiming to deepen the understanding of cortical dynamics. The findings suggest improved data reliability for closed-loop neurostimulation applications, potentially expanding the utility of TMS-EEG in diagnosing and treating neurological conditions through more precise monitoring and intervention strategies.
cs.AI updates on arXiv.org