IntroductionThis study explores the feasibility of implementing a tractography-based navigated transcranial magnetic stimulation (nTMS) language mapping protocol targeting cortical terminations of the arcuate fasciculus (AF). We compared the results and distribution of errors from the new protocol to an established perisylvian nTMS protocol that stimulated without any specific targeting over the entire perisylvian cortex.MethodsSixty right-handed patients with language-eloquent brain tumors were examined in this study with one half of the cohort receiving the tractographybased protocol and the other half receiving the perisylvian protocol. Probabilistic tractography using MRtrix3 was performed for patients in the tractography-based group to identify the AF’s cortical endpoints. nTMS mappings were performed and resulting language errors were classified into five psycholinguistic groups.ResultsTractography and nTMS were successfully performed in all patients. The tractogram-based group showed a significantly higher median overall ER than the perisylvian group (3.8% vs. 2.9% p <.05). The median ER without hesitation errors in the tractogram-based group was also significantly higher than the perisylvian group (2.0% vs. 1.4%, p <.05). The ERs by error type showed no significant differences between protocols except in the no response ER, with a higher median ER in the tractogram-based group (0.4% vs. 0%, p <.05). Analysis of ERs based on the Corina cortical parcellation system showed especially high nTMS ERs over the posterior middle temporal gyrus (pMTG) in the perisylvian protocol and high ERs over the middle and ventral postcentral gyrus (vPoG), the opercular inferior frontal gyrus (opIFG) and the ventral precentral gyrus (vPrG) in the tractography-based protocol.DiscussionBy considering the white matter anatomy and performing nTMS on the cortical endpoints of the AF, the efficacy of nTMS in disrupting patients’ object naming abilities was increased. The newly introduced method showed proof of concept and resulted in AF-specific ERs and noninvasive cortical language maps, which could be applied to additional fiber bundles related to the language network in future nTMS studies.

White matter impairments caused by gliomas can lead to functional disorders. In this study, we predicted aphasia in patients with gliomas infiltrating the language network using machine learning methods. We included 78 patients with left-hemispheric perisylvian gliomas. Aphasia was graded preoperatively using the Aachen aphasia test (AAT). Subsequently, we created bundle segmentations based on automatically generated tract orientation mappings using TractSeg. To prepare the input for the support vector machine (SVM), we first preselected aphasia-related fiber bundles based on the associations between relative tract volumes and AAT subtests. In addition, diffusion magnetic resonance imaging (dMRI)-based metrics [axial diffusivity (AD), apparent diffusion coefficient (ADC), fractional anisotropy (FA), and radial diffusivity (RD)] were extracted within the fiber bundles' masks with their mean, standard deviation, kurtosis, and skewness values. Our model consisted of random forest-based feature selection followed by an SVM. The best model performance achieved 81% accuracy (specificity = 85%, sensitivity = 73%, and AUC = 85%) using dMRI-based features, demographics, tumor WHO grade, tumor location, and relative tract volumes. The most effective features resulted from the arcuate fasciculus (AF), middle longitudinal fasciculus (MLF), and inferior fronto-occipital fasciculus (IFOF). The most effective dMRI-based metrics were FA, ADC, and AD. We achieved a prediction of aphasia using dMRI-based features and demonstrated that AF, IFOF, and MLF were the most important fiber bundles for predicting aphasia in this cohort.