Back

Validating models of TMS effects with concurrent TMS/fMRI

Petrov, P. I.; Vink, J.; Mandija, S.; van den Berg, N.; Dijkhuizen, R.; Neggers, S. F. W.

2023-07-07 neuroscience
10.1101/2023.07.05.547836 bioRxiv
Show abstract

The exact dose and spatial pattern of brain activation induced by transcranial magnetic stimulation (TMS) is of importance to all applications of TMS, both clinical and for investigational use. How TMS coils induce electrical fields has been investigated and validated by several groups, but models of how evoked currents interact with neuronal tissue and where activation is induced have hardly been validated empirically. Here we propose a detailed model of TMS induced currents in the brain building on our previous work on different TMS coil models, taking into account various electromagnetic head tissue properties with finite element modeling and several competing macroscopic models of how brain activation is achieved by the TMS induced electrical field. Importantly, we validate these models using TMS coil discharges administered to the brain of 6 healthy volunteers inside a 3T MRI scanner, while acquiring functional MRI scans at the same time. We conclude that the mere magnitude of the electric field in the cortical gray matter tissue near the stimulation site provides the best metric to predict neuronal activation as measured by BOLD fMRI, whereas models of brain activation taking into account orientation of E-fields with respect to the cortical columns and layers do not (yet) yield better predictions of neuronal activation induced by TMS. This report is relevant for neuronavigated TMS approaches attempting to take into account evoked E-fields and neuronal activation to improve precision of TMS administration for clinical and investigational purposes.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.