Artikel
Using resting-state-fMRI to identify language-related areas in tumor patients – A feasibility study
Suche in Medline nach
Autoren
-
Christian Ott
- Klinik für Neurochirurgie, Universitätsklinikum Regensburg
-
Markus Goldhacker
- Department of Experimental Psychology, Universität Regensburg
-
Katharina Rosengarth
- Klinik für Neurochirurgie, Universitätsklinikum Regensburg; Department of Experimental Psychology, Universität Regensburg
-
Christian Doenitz
- Klinik für Neurochirurgie, Universitätsklinikum Regensburg
-
Elmar Lang
- Institut für Biophysik, Universität Regensburg
-
Mark Greenlee
- Department of Experimental Psychology, Universität Regensburg
-
Alexander Brawanski
- Klinik für Neurochirurgie, Universitätsklinikum Regensburg
| Veröffentlicht: | 2. Juni 2015 |
|---|
Gliederung
Text
Objective: Task-based functional MRI (tb-fMRI) has been successfully used to identify language-related areas in patients suffering from lesions close to those areas. Patients must be able to actively participate in such examinations. In contrast, there is no need for an active participation of the patient in resting-state fMRI (rs-fMRI). Exploiting this fact, the major question of our study is, whether it is possible to identify language-related areas in patients suffering from a lesion close to those areas using rs-fMRI with comparable accuracy to tb-fMRI in order to use results from rs-fMRI for surgical navigation.
Method: 6 patients suffering from a lesion close or adjacent to language related areas were included in this study. All of the patients underwent tb-fMRI for speech recognition and generation, and rs-fMRI, which was analyzed by independent component analysis (ICA). ICA was performed in 9 runs for each patient (20 to 100 components). Correlation analysis between the tb-fMRI-contrast and rs-fMRI-components was conducted. For each patient the component – out of all ICA runs – highest correlating with the tb-fMRI-contrast was selected as a candidate for representing language related areas.
Results: Preliminary results showed that in 5 out of 6 patients the correlation coefficient of the maximally correlating component of the ICA of the rs-fMRI with the tb-fMRI-contrast was relatively large. Visual investigation of those components was promising concerning the activation around tumor areas and their appearance as being language-related.
Conclusions: Tb-fMRIs are stretched to their limits by patients, who are not able to actively participate during an examination (e.g. cognitive disorders, incompliance, sedation, preexistiting neurological impairment). Rs-fMRI could be an alternative tool for the identification of the eloquent cortex in those patients. The major problem is the identification of the correct component and thereby the correct (language) network. In this study we could show that it is possible to identify language-related areas in tumor patients using rs-fMRI comparably to tb-fMRI. In further studies we aim to show that rs-fMRI can serve as the sole surgical planning- and navigation-tool.
