Pre-surgical Functional Brain Mapping

Among the issues that MEG/MSI, as well as all the other functional imaging techniques are called to address, is the identification of the brain regions mediating sensation, movement, language and memory. Advance knowledge of these regions facilitates surgical planning and reduces morbidity associated with resection of eloquent cortex, especially in cases of epilepsy surgery. Such knowledge was typically sought through invasive means such as the Wada procedure and direct cortical stimulation either intraoperatively or extraoperatively via implanted electrodes or subdural electrode grids. These invasive methods of brain mapping are now supplemented or replaced by non-invasive functional imaging techniques such as MEG/MSI and functional MRI.

brain scan image
Localization of somatosensory (left) and language (right) cortices relative to interictal epileptiform loci, using MEG/MSI.

TMS is being used to preoperatively map the location and extent of the motor cortex when a brain tumor or the epileptogenic focus is in close proximity to the motor cortex or the eloquent cortex, and surgery is considered. Preoperative motor mapping with TMS is becoming an important tool available to neurologists and neurosurgeons in planning surgeries. The TMS lab at Le Bonheur Children's Hospital has been operational since July 2012. In the figures below, Cases 1 and 2 are examples of presurgical mapping performed at Le Bonheur Children's Hospital. In addition, these Cases 3 and 4 demonstrate the utility of TMS in studying plasticity in the motor system.

Case 1: A 52 year old female diagnosed with a tumor in the left motor cortex. Motor mapping indicated that the mouth motor cortex was in the vicinity of the tumor while the hand motor cortex was not (left panel). The tumor was found to be near speech production areas (Broca's area) (right panel).

Case 2: A 21 year old male whose tumor in the right frontal lobe was removed, continued to have seizures. His seizure focus was in the margins of the resection cavity. Motor mapping indicted that the hand motor cortex was along the posterior margin of the cavity.

Case 3: A 6 year old female had onset of seizures at 5 months of age and diagnosed with right hemisphere dysplasia and developmental delay. Patient underwent a right functional hemispherectomy at age 3. Presently patient continues to have seizures characterized by bilateral arm movements. The locations of primary motor cortex that innervate the right adductor pollicis brevis (APB) and adductor digiti minimi (ADM) muscles are normally located along the left precentral gyrus (blue and purple circles respectively). The cortical location of cortex that innervates the left brachioradialis (Brac) muscle is located in the left post central gyrus, (shown in yellow). The MEPs are shown in the right panel.

Case 4: A 6 year old male sustained a traumatic brain injury at 15 months of age, resulting in extensive encephalomalacia of the right cerebral hemisphere with secondary mental retardation and refractory partial seizure disorder. The patient currently has a limited use of his left arm, with no signs of muscular wasting. The locations of primary motor cortex that innervate the right hand muscles are located along the left precentral gyrus and post central gyrus (orange pegs). The cortical locations of cortex that elicited MEPs in both hands (indicating innervation of the left hand as well) are located in the left post central gyrus, (yellow pegs).


To reduce the risk of neurological impairment after surgery for epilepsy or tumors, it is often necessary to map cortical function in each patient due to individual variability in the functional anatomy and possible lesion-induced functional reorganization, particularly with respect to language. Cortical stimulation mapping (CSM) is usually utilized for functional mapping during which an electrical current is sequentially passed through adjacent subdural electrodes and the language cortex is identified by inhibition of language function in the stimulated electrodes. The CSM is accepted as the gold standard for functional mapping, though it has several limitations. For example, CSM can produce after-discharges and electrically-induced seizures that put the patient at risk and make additional immediate testing unpredictable or even impossible. Furthermore, CSM is time-consuming and requires patient cooperation that makes functional mapping in young, uncooperative, and developmentally delayed patients quite challenging.


It has been shown that enhancement of the power of ECoG recording in the high gamma (HG) frequency range (> 50 Hz) is a reliable marker of local cortical activation in sensorimotor and higher functions, such as face perception, working memory, and language. In particular, it has been found that enhancement of the power of ECoG recording in the HG frequency range is a reliable marker providing functional language mapping of high spatial and temporal resolution. Results of our studies suggest that language mapping on the basis of high gamma ECoG can provide important additional information, therefore, this methods can be used in conjunction with CSM or as an alternative, when the latter is deemed impractical.


Using CSM and subdural ECoG recordings, we explored the organization of expressive language cortex in two patients undergoing epilepsy surgery evaluation (Babajani-Feremi et al, 2014). Fig. 1 shows the locations of the subdural electrodes for two patients. Fig. 2 shows HG activity during the object naming task. As shown in Fig. 2, contacts 32 and 24 in Patient 1 were the only electrodes that recorded significant HG activity during the object naming task. Significant enhancement of the HG activity in contact 32 was started at approximately 500 ms following stimulus onset and was sustained during the period that overt articulation was taking place. These observations suggest that contact 32 in Patient 1 overlays the speech production area. HG activity at contact 24 in Patient 1 was only present in the latency corresponding to overt articulation and there was not significant before articulation and during the presentation of the pictures. These observations suggest that contact 24 in Patient 1 overlays the motor speech areas. In fact, the CSM results confirmed this hypothesis in that it revealed that stimulation of contact 32 resulted in complete speech arrest and stimulation through contact 24 resulted on contraction of the tongue.


Contacts 15 and 7 in Patient 2 were the only ones that demonstrated significant HG activity during the object naming task (see Fig. 2). Significant enhancement of HG in contact 15 was started before overt articulation, approximately 500 ms following presentation of the pictures. Significant HG activity at contact 7 was only observed during overt articulation and there was no significant HG activity before articulation. Temporal characteristic of HG activity in the object naming task suggest that contacts 15 and 7 in Patient 2 overlays, most likely, speech production and motor speech areas, respectively. In fact, the CSM results confirmed these findings.


In another study, the pre-surgical language mapping using CSM was compared with that using HG ECoG, fMRI, and TMS for a patient with epilepsy (i.e. Patient 3, male, 18 years old) who underwent surgery, and the results are shown in Fig. 3. Significant enhancement of power of high gamma during an object naming task was observed in two CSM-positive electrodes (31 and 32). There were also six false positive electrodes (5, 6, 26, 27, 33, and 34) showing significant high gamma activity but stimulation of which did not result in speech arrest. In the object naming task, patients were instructed to start articulation 1.5 second after stimulus onset. The significant high gamma activity in electrodes 26, 27, 33, and 34 was observed about 2 second after stimulus onset during the articulation but not before the articulation. The CSM results revealed that stimulation of the bipolar electrodes 33 and 34 caused tongue contractions. Considering results of CSM and hgECoG, the neighboring electrodes 26, 27, 33, and 34 most likely lay on the motor area of language. Fig. 3-c shows the fMRI language mapping in this patient in which several cortical areas, including Broca’s area (BA 44/45), the premotor cortex (BA 6), the superior temporal gyrus (Wernicke’s area), and visual cortex were active during the object naming task. The fMRI activation covers two CSM-positive electrodes which correspond to two true positive electrodes and zero false negative electrode and, consequently, sensitivity of 100% for this patient. Four electrodes laid on fMRI activation but stimulation of which did not cause speech arrest, and thus these electrodes were considered as false positive. TMS language mapping in this patient is shown in Fig. 3-d in which the expressive language cortex were identified by six sites, marked as white stars. Despite the scattered fMRI activation in this patient, the TMS language map in this patient was focal and its center located in the Broca’s area (BA 44) close to the CSM-positive electrodes.


 Fig. 1

Fig. 1. Subdural electrode locations and cytoarchitectonic maps of Broca’s areas (BA 44/45) are shown on a rendering of the standard Montreal Neurological Institute (MNI) brain atlas and native (subject) space. The red and blue maps show BA 44 and BA 45 areas, respectively. Subdural electrodes for Patient 1 consisted of a modified 32-channel grid placed over the lateral temporal region and a 4-contact inferior frontal strip (IF1-IF4). Subdural electrodes for Patient 2 consisted of a modified 64-channel grid placed over the frontal region. Interruption of expressive speech was observed during stimulation of pair electrodes (31-32) and (14-15), marked by black lines, of Patient 1 and Patient 2, respectively. Electrical stimulation of pair electrodes (23-24), marked by a magenta line, resulted in tongue contraction in Patient 1.


fig 2

Fig. 2. Time-frequency (TF) analysis of recordings during the object naming task in a subset of subdural electrodes. Power in every time-frequency bin was compared to the mean power in the baseline (i.e., [-0.5 -0.1] sec) to calculate a t-value across all trials. The top and bottom parts of this figure show the TF results for Patient 1 and Patient 2, respectively. The title of each subplot shows the subdural electrode number (see Fig. 1 for locations of the electrodes). Significant enhancement of the high gamma is only present in electrodes 24 and 32 in Patient 1, and electrodes 7 and 15 in Patient 2. Time 0 indicates the onset of the visual stimulus.


Fig. 3. Language mapping using CSM is compared with that using hgECoG, fMRI, and TMS in a representative patient (i.e., Patient 3), and the results are shown in standard MNI atlas. Subdural electrodes which were not stimulated during CSM are shown in magenta, and the rest of electrodes were identified as true negative (TN), true positive (TP), or false positive (FP). This patient did not have any false negative electrode. Significant postoperative language deficits were observed in this patient. The black dashed line demarcates border of the resected anterior temporal tip. The resected zone included two high gamma positive electrodes and parts of the language-specific cortex identified using TMS. (a) The relative change of power of high gamma with respect to a baseline ([-0.5 -0.1] s), averaged across the frequency range of [85 90] Hz and time interval of [0.60 0.65] second after stimulus onset, during the overt object naming task is calculated and a render of that on the MNI cortical surface is shown. (b) The relative change of power of high gamma with respect to a baseline ([-0.5 -0.1] s) in a true positive (hgECoG-positive and CSM-positive) electrode (i.e. electrode 31 in (a)) during the overt object naming task is shown. The cartoon at the bottom of the time-frequency plot shows timing of the object naming task in which objects was presented visually for 1.5 sec and after that patients articulated the name of objects. (c) Projection of the significant fMRI activity (P < 0.05 with FWE correction) on cortical surface is shown. (d)The expressive language mapping using TMS resulted in disruption of speech production in six sites, marked as white stars. Based on the locations of these sites, a smeared image was calculated and voxels in this image which survived at a threshold of 50% were projected into the cortical surface.


The following documents are in PDF format and require Adobe Reader, which is a free download.

Contact Us

Clinical Neurosciences
Department of Pediatrics
UTHSC College of Medicine
The Neuroscience Institute
Le Bonheur Children's Hospital
51 N. Dunlap Street P320
Memphis, TN 38105