Artikel
Cortical language mapping by rTMS compared to fMRI in healthy volunteers
Suche in Medline nach
Autoren
Veröffentlicht: | 8. Juni 2016 |
---|
Gliederung
Text
Objective: While functional magnetic resonance imaging (fMRI) was considered to be the standard for non-invasive language mapping for a long time, repetitive navigated transcranial magnetic stimulation (rTMS) is an upcoming method. However, most current language models refer to fMRI language mapping data. We therefore compared both modalities for language mapping in healthy subjects.
Method: We performed rTMS and fMRI language mapping in 40 healthy subjects. After assigning language-positive regions to a cortical parcellation system (CPS), we calculated rTMS error rates (ER = errors per stimulations) and fMRI activation rates (AR = blood-oxygen-level-dependent-positive regions per patients) for each CPS region. We defined different error rate thresholds (ERT = ER at which a CPS region was defined as language-positive in terms of rTMS), and used Cohen’s kappa coefficient for the comparison of the two techniques.
Results: The overall kappa for all regions was 0.05 ± 0.06 (mean ± SD). We found the best agreement for both techniques when using the 2-out-of-3 rule (CPS region was defined as language-positive in terms of rTMS if at least 2 out of 3 stimulations led to a language error). However, kappa value for this agreement was only 0.24, describing a fair agreement.
Conclusions: Mapping data by fMRI and rTMS differed significantly and we need to be careful to call one of the two techniques as the true one. However, current language models referring to fMRI data should be reconsidered, and new models should combine data of both techniques.
Note: Sebastian Ille and Nico Sollmann contributed equally.