Identifying site- and stimulation-specific TMS-evoked EEG potentials using a quantitative cosine similarity metric

Abstract

The ability to interpret transcranial magnetic stimulation (TMS)-evoked electroencephalography (EEG) potentials (TEPs) is limited by artifacts, such as auditory evoked responses produced by discharge of the TMS coil. TEPs generated from direct cortical stimulation should vary in their topographical activity pattern according to stimulation site and differ from responses to sham stimulation. Responses that do not show these effects are likely to be artifactual. In 20 healthy volunteers, we delivered active and sham TMS to the right prefrontal, left primary motor, and left posterior parietal cortex and compared the waveform similarity of TEPs between stimulation sites and active and sham TMS using a cosine similaritybased analysis method. We identified epochs after the stimulus when the spatial pattern of TMS-evoked activation showed greater than random similarity between stimulation sites and sham vs. active TMS, indicating the presence of a dominant artifact. To do this, we binarized the derivatives of the TEPs recorded from 30 EEG channels and calculated cosine similarity between conditions at each time point with millisecond resolution. Only TEP components occurring before approximately 80 ms differed across stimulation sites and between active and sham, indicating site and condition-specific responses. We therefore conclude that, in the absence of noise masking or other measures to decrease neural artifact, TEP components before about 80 ms can be safely interpreted as stimulation locationspecific responses to TMS, but components beyond this latency should be interpreted with caution due to high similarity in their topographical activity pattern.

Probing regional cortical excitability via input–output properties using transcranial magnetic stimulation and electroencephalography coupling

“…The quality of regression of the realistic sham conditions on early components never significantly differed from noise, as quantified with the classical sham procedure used in the first experiment. This is in line with recent findings suggesting that only late components appear to contain significant PEPs (Biabani et al, ; Freedberg et al, ). In contrast, recruitment curves drawn from active stimulation differed from noise and showed different patterns across sites, possibly revealing different input–output properties of the cortical tissue.…”

Single-Pulse TMS to the Temporo-Occipital and Dorsolateral Prefrontal Cortex Evokes Lateralized Long Latency EEG Responses at the Stimulation Site

“…These findings are consistent with the notion that the N100 is sitespecific and reflects local intracortical excitability-inhibition networks in the targeted brain region. By contrast, other studies found the TMS-evoked N100 to be uniform across several different stimulated brain areas with a stereotypical symmetrical distribution over the vertex irrespective of the targeted cortex region, therefore interpreting it as an unspecific response representing global properties of the brain or even an artifact (Du et al, 2017;Freedberg et al, 2020). In order to use TEPs in neuropsychiatric research and to adequately translate findings into applications as a neurostimulation biomarker, it is crucial to determine which TEP components reflect local cortical properties evoked by direct transcranial effects.…”

A Structured ICA-based Process for Removing Auditory Evoked Potentials Reveals TMS-evoked Potentials and TMS-modulated Oscillations

“…The TEP re ects TMS-evoked brain activation [18][19] that can be speci c to stimulation site [9,12,[20][21][22], and reproducible [12,20,23], particularly at earlier latencies. However, the speci city of longer latency responses is a matter of debate [15,[24][25] due to high correlation between TEPs and sensory responses to TMS by 60-80 ms [16-17,20] and high correlation between active and sham stimulation starting as early as 58 ms [15,17].…”