Secondary Logo

Journal Logo

Language Mapping With fMRI

Current Standards and Reproducibility

Agarwal, Shruti PhD*; Sair, Haris I. MD*; Gujar, Sachin MBBS*; Pillai, Jay J. MD*,†

Topics in Magnetic Resonance Imaging: August 2019 - Volume 28 - Issue 4 - p 225–233
doi: 10.1097/RMR.0000000000000216
Review Articles

Clinical use of blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) is a relatively new phenomenon, with only about 3 decades of collective experience. Nevertheless, task-based BOLD fMRI has been widely accepted for presurgical planning, over traditional methods, which are invasive and at times perilous. Many studies have demonstrated the ability of BOLD fMRI to make substantial clinical impact with respect to surgical planning and preoperative risk assessment, especially to localize the eloquent motor and visual areas. Reproducibility and repeatability of language fMRI are important in the assessment of its clinical usefulness. There are national efforts currently underway to standardize language fMRI. The American Society of Functional Neuroradiology (ASFNR) has recently provided guidelines on fMRI paradigm algorithms for presurgical language assessment for language lateralization and localization. In this review article, we provide a comprehensive overview of current standards of language fMRI mapping and its reproducibility.

*Division of Neuroradiology, Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD

Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD.

Address correspondence to Jay J. Pillai, MD, Division of Neuroradiology, Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins Hospital, Phipps B-100, 1800 Orleans Street, Baltimore, MD 21287 (e-mail:

Received 3 June, 2019

Accepted 10 June, 2019

The authors report no conflicts of interest

Clinical use of blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) is a relatively new phenomenon, with only about 3 decades of collective experience. Nevertheless, task-based BOLD fMRI has been widely accepted for presurgical planning in patients with brain tumors and other focal brain lesions such as brain neoplasms, arteriovenous malformations and other vascular malformations, cortical dysplasias, and other epileptogenic lesions.1–4 Preoperative fMRI has been very helpful in assisting neurosurgeons in making decisions of whether or not to attempt surgical resection of the lesion and in selection of patients for asleep vs awake craniotomy. The preoperative detection of functional areas around a lesion can guide the intraoperative cortical stimulation (ICS) mapping, thus reducing the total surgical time and planning the safest surgical trajectory. BOLD fMRI has recently been used in patients with trauma, vascular diseases, inflammations, multiple sclerosis, Alzheimer disease, developmental disorders, learning disabilities, and many other conditions, although most of these applications are still considered to be in the research realm.

Back to Top | Article Outline


For a long time, it was mistakenly believed that cerebral hemodynamics were solely controlled by the brain's short-term metabolic requirements. Contrary to this popular belief, in 1986, Fox and Raichle5 demonstrated using O15-labeled radiotracers positron emission tomography (PET), the independence between brain blood flow and oxygen demand. In that study, they found that regional cerebral blood flow (CBF) increases during cerebral activation; however, the cerebral metabolic rate of oxygen consumption (CMRO2) does not increase proportionally. Consequently, more fresh (oxygenated) blood is supplied to that region of functioning brain than is required for its immediate metabolic needs. Effectively, this results in decreased relative concentration of deoxyhemoglobin compared with oxyhemoglobin. In 1992, Ogawa et al6 exploited different magnetic properties of oxygenated and deoxygenated blood to detect changes in regional CBF. This property, popularly known as BOLD (blood oxygen level dependent), provides contrast for the most commonly performed clinical functional MRI (fMRI). A decrease in the relative concentration of deoxyhemoglobin in active cortex reduces the T2/T2* shortening effects of deoxyhemoglobin with resultant net increase in the BOLD signal in activated areas.

Back to Top | Article Outline

Imaging Strategies and Tradeoffs

fMRI necessitates image acquisition methods, which are sensitive to changes in T2* and T2, have sufficient spatial resolution to cover the entire brain, and have sufficient temporal resolution to detect changes in BOLD signal associated with specifics tasks.7,8 Acquisition strategies are largely based on a trade-off between signal-to-noise ratio (SNR), spatial resolution, temporal resolution, and motion artifacts.

T2*-weighted gradient echo (GRE) sequences9,10 are the most commonly used BOLD sequences at fields of 3T or below. The high sampling rates necessary for fMRI are achieved utilizing “single-shot” sequences, which make use of fast gradient switching technologies allowing for acquisition of the data for an entire slice in one readout window after one excitation. Echo-planar imaging (EPI)11,12 and spiral imaging13,14 are the most widely used single-shot imaging sequences. Recently developed “parallel imaging (PI)” techniques reduce the readout time by acquiring data from multiple coils.15,16 For BOLD acquisitions at 1.5T and 3T, PI is not preferable because of high SNR loss at even the lowest acceleration factor.17 A recent development is multiband parallel imaging technique. In multiband EPI, multiple 2D slices are simultaneously excited (multiband excitation), and individual slices are reconstructed.18 Multiband is not hindered by the SNR reduction associated with reduced sampling in traditional parallel imaging19; however, interslice leakage artifact occurs due to incomplete separation of slice signals over time.20 Artifacts such as noise, motion, aliasing, chemical shift, Gibbs, susceptibility, RF-interference, and so on, which are usually noticed in conventional MR imaging, also occur in PI with slightly different appearances because of the complex PI reconstruction process.21

Back to Top | Article Outline

Language fMRI Paradigm Designs

Appropriately designing the stimulus paradigm is as important as choosing imaging parameters. A valid stimulus paradigm should ideally localize only the areas pertaining to the considered function of interest. The location and extent of activated areas should not vary significantly in different trials during the same or different sessions, that is, the paradigm should be reproducible.

During an fMRI scan session, the brain is repeatedly imaged while the subject is resting or performing the control task and while the subject is presented with a stimulus or is performing an active task. The paradigm timing depends on tradeoff between the level of attention of the subject during the task, avoiding motion and acquiring enough data to provide statistically significant mapping results. For presurgical planning, block designs are most commonly used that typically last between 2 and 4 minutes with epochs of 15 and 30 seconds. In the block design, there are regular epochs of stimulus and rest, labeled “on” (active) and “off” (control). The resulting difference between the signal acquired during the “on” period and the “off” period is the activation of interest. The “off” condition is not always the absence of any activity, but may also consist of another task that differs from the primary task performed during the “on” period and does not activate the functional network of interest.

A large variety of paradigms have been reported in the literature for mapping the language network with fMRI.3,22–33 These paradigms can be divided into 3 main categories: expressive, receptive, and semantic. Expressive paradigms are designed to elicit language productive areas, particularly Broca area (BA) in the dominant hemisphere. Frontal brain regions, including precentral gyrus, superior frontal gyrus, middle frontal gyrus, inferior frontal gyrus (including pars opercularis, pars triangularis and pars orbitalis), and supplementary motor area, are expected to see activation during expressive language task performance. Verbal fluency tasks are very frequently used for presurgical expressive language brain mapping. Receptive paradigms are designed to activate the eloquent cortex areas related to comprehension of language, particularly Wernicke area (WA) in the dominant hemisphere. Temporal brain regions, including fusiform gyrus, temporal lobes (superior temporal gyrus, middle temporal gyrus, inferior temporal gyrus, temporal poles-superior, poles-mid), and parietal brain regions including supramarginal gyrus and angular gyrus are expected to see activation during receptive language task performance. Receptive paradigms include mainly language reading or listening comprehension tasks. Semantic paradigms34 are more general purpose for language mapping than expressive and receptive paradigms. They are designed to be useful for language lateralization and localization of inferior frontal, posterior temporal, and inferior parietal speech areas in the dominant cerebral hemisphere. Figure 1 displays general regions of the brain where language related functional areas may be located in the Frontal, Temporal and Parietal lobes.



Benjamin et al23 categorized 6 critical regions for presurgical language mapping: BA within the inferior frontal gyrus (pars opercularis and pars triangularis)35; Exner area (“graphemic motor frontal area”) in the posterior region of the middle frontal gyrus, critically involved in transforming phonological representations of words into motor commands for producing their written forms36; Supplementary speech area (“pre-supplementary motor area,” speech component; pre-SMA) in the posterior superior and medial frontal cortex, critically engaged in initiating and sequencing motor movements for speech; Angular gyrus, particularly involved in reading and transitioning between written and spoken forms of language; WA in temporal lobe gyri (specifically, superior and middle temporal gyri) with multiple varied definitions in use however largely representing receptive language function35,37; Basal Temporal language area, critical to language in the basal temporal region,38 a region thought to link semantics with names.

The American Society of Functional Neuroradiology (ASFNR) hosts presentation files and parameters of the most common paradigms that are freely available for use on their website ( Additional unique paradigms utilized by institutions across the country are also described in a document hosted by the ASFNR website ( (1) Auditory Responsive Naming Task (Aural/Visual) in which the patient is asked to read a sentence and push a button corresponding to the correct choice (visual presentation) and silently generate a word that fits the description (aural presentation). The baseline is a resting baseline for the aural presentation and a nonsense baseline for the visual presentation. The task has been used to localize WA and may be a more sensitive measure than the typical object naming during electrocorticography.39 (2) Semantic Decision Task (Visual) in which word pairs (eg, “fruit” and “apple”) are presented and the patient is asked to push a button when pairings are correct. The baseline is a visual presentation of pair of lines and patient is asked to push the button when the 2 sets of lines are identical. This semantic decision paradigm has been used for language lateralization and localization of inferior frontal, posterior temporal, and inferior parietal speech areas in the dominant cerebral hemisphere. (3) Text Reading versus Non-Linguistic Symbols40 in which the patient is instructed to concentrate on reading and comprehending text during the active block and merely attend to the symbols during the control block. This task has been used to localize language comprehension areas in the dominant hemisphere region of posterior superior temporal gyrus/parietal angular gyrus (WA). (4) The Visual Language Comprehension in which sentence-question pairs are presented and patient is asked to push a 2-button finger switch held in the dominant hand for a YES/NO answer. (5) Silent Verb Generation Task, in which the patient is asked to generate verbs following presentation of a noun; baseline is visual fixation. (6) Word Listening Task41 for which the patient hears words (nouns), during which the patient is supposed to repeat the words silently.

Table 1 provides a comprehensive list of fMRI paradigms, which are currently used to activate language brain areas at the Johns Hopkins Hospital (JHH). Figure 2 provides samples of control block and active block of language fMRI paradigm designs which are currently used at the JHH.42–51





Back to Top | Article Outline

Monitoring and Processing

A comprehensive pre-scan training session outside the MRI scanner should be performed to ensure full patient understanding of task instructions and confirmation of each patient's ability to adequately perform the tasks. Each patient's task performance should be monitored during the scan via real-time fMRI for assessment of activation, bulk head motion, and physiologic noise.

Once the data are obtained, raw data are corrected for slice timing due to interleaved EPI scans, followed by realignment to correct for motion of the subject during the functional scans utilizing rigid body translation and rotation transformation. Thereafter, as a pre-processing step, spatial smoothing of the images with a Gaussian kernel should be performed on raw data.

For generating activation maps from the pre-processed data, regression analysis is performed by fitting the observed fMRI signal time course of each voxel to a theoretical expected time course generated by a standard hemodynamic response function (HRF). The theoretical HRF is a mathematical expression of the mechanism behind BOLD fMRI. Due to an increase in neuronal activity, fMRI signal initially decreases because the active neurons consume oxygen thereby increasing the relative level of deoxyhemoglobin in the blood.52,53 This decrease, however, is tiny and is not always found.54 After an initial undershoot, fMRI signal reaches its maximum in approximately 6 seconds.52–55 This increase is due to oversupply of oxygen-rich blood, which causes a large decrease in the relative level of deoxyhemoglobin, which in turn results in increased BOLD signal. Finally, the level of deoxyhemoglobin slowly returns to normal after an initial undershoot in approximately 24 seconds52 and the fMRI signal decays until it has reached its original baseline level. This HRF is then convoluted with a condition box-car function having a value of 1 corresponding to active blocks of the fMRI paradigm and 0 for control blocks. T-contrast maps are then obtained to analyze the contrast between the language activation and baseline conditions. Figure 3 display suprathreshold language activation maps overlaid on anatomical T1 MPRAGE obtained from SC and SWG tasks performed in the same scan session by a right-handed patient with left medial temporal lobe lesion.



Back to Top | Article Outline

Lateralization and Localization

Language function is known to display unilateral hemispheric dominance.56 Presurgical fMRI mapping for the assessment of language areas in patients with focal brain lesions has been used to determine hemispheric language dominance, thus replacing traditional invasive approach of intracarotid amobarbital procedure57 also known as the “Wada Test,” which involves arterial catheterization. Many studies have clinically validated hemispheric language lateralization obtained from language fMRI mapping through comparison with hemispheric dominance obtained from Wada testing; concordance between lateralization indices (LIs) from these 2 techniques has generally been in the range of 90% to 100%.58–63

In a recent review article, Bradshaw et al64 evaluated various methods used in fMRI studies since the year 2000 for quantifying laterality. Laterality Index65 as a means to evaluate relative extents of activation in the left and right hemispheres has been used widely to assess language hemispheric dominance.42,66–68 LI is computed as (L-R)/(L+R), where L refers to activations in the left hemisphere and R refers to activations in the right hemisphere. This formula yields values between −1 and +1, which are positive for left hemispheric dominance (LI ≥ 0.2), negative (LI ≤ −0.2) for right hemispheric dominance, and bilateral when (−0.2<LI<0.2).69 LI can be determined by comparing all positively task correlated voxels between the left and right hemisphere, referred as “Threshold-Independent LI”. When LI is computed only between supra-threshold voxels in language activation map, it is referred to as “Threshold-dependent LI.”42 In their review article, Bradshaw et al64 suggested threshold-independent LI methods for assessing heterogeneity of language laterality across multiple regions of interests and various language tasks. Pillai et al42 compared threshold-dependent versus threshold-independent techniques in a large series of brain tumor patients undergoing presurgical language mapping. Their findings suggested that expressive tasks provided the best hemispheric language lateralization based on concordant threshold-dependent and threshold-independent analyses and receptive tasks were less effective for language lateralization.

Language presurgical fMRI mapping not only localizes critical areas for language processing but also identifies other less critical areas of the language network. Various studies have been performed to validate language fMRI by comparing the locations of activation found during fMRI with the language functional mapping obtained from Ojemann Stimulator70 during awake craniotomy.43,71,72 As not all patients are candidates for awake ICS mapping, preoperative fMRI often becomes essential for surgical planning. Many studies have also demonstrated the ability of fMRI to make a substantial clinical impact, particularly with respect to surgical planning and preoperative risk assessment.73–83

Back to Top | Article Outline

Reproducibility and Repeatability

Reproducibility and repeatability of language fMRI is an important measurement in the assessment of its clinical usefulness. fMRI tasks are often repeated to confirm both across-subject reproducibility and within-subject repeatability of language lateralization and localization.

Many studies have evaluated across-subject reproducibility of language activations to assess the reliability of fMRI paradigms by calculating the correlation coefficients as similarity metric between the activations obtained from the same tasks performed in 2 or multiple sessions by same participants.84–90 Another prominent measure of similarity metric is the Dice coefficient.91 Dice coefficient and other reliability metrics are used to quantify test-retest variability of activation clusters across sessions and across subjects. Several reliability studies of language mapping paradigms inform about the consistency of activation between subjects.92–94

There are national efforts currently underway to standardize language fMRI. In a recent white paper, the ASFNR provided guidelines on fMRI paradigm for presurgical language assessment.95 In the ASFNR guidelines, the sentence completion and silent word generation tasks were recommended as primary tasks to be used for effective language cortical localization and hemispheric lateralization/dominance determination. The ASFNR conducted a poll among its entire membership on the number of language-assessment fMRIs per month that were performed at their various institutions to determine the most commonly used language paradigms. Results of the ASFNR poll produced 3 tiers of commonly used paradigms across the nation. Silent Word Generation and Sentence Completion stood out as the most frequently employed tasks followed by Verb Generation, Object Naming, Rhyming, and Reading Comprehension in the second-tier group. On the basis of the poll results and literature review, the ASFNR task force recommended language paradigm algorithms for both adults and pediatric subjects. As language can be represented across phonologic, semantic, receptive, expressive, and so on domains and one task cannot simultaneously activate all of these aspects, multiple tasks are recommended to provide a more sensitive and specific map of language function that will aid in surgical planning. The default adult algorithm includes SC, SWG, and Rhyming. The radiologist may choose to drop Rhyming and repeat the SC (most appropriate task to repeat) or SWG task. Radiologist may choose either the Object Naming (which is an expressive task) or Passive Story Listening (which is a receptive task) task when patient has difficulty performing SC/SWG tasks. The default pediatric algorithm includes SC, Rhyming, and Antonym Generation (AG).

In a recent paper, Agarwal et al96 demonstrated a very high degree of repeatability at the single-subject level within single scan sessions of language mapping with sentence completion and silent word generation tasks in a large cohort of patients (37 right-handed patients for SC task and 78 right-handed patients for SWG task) undergoing presurgical fMRI. They investigated the laterality index and center of mass (COM) of holohemispheric (ie, individual left and right hemispheres excluding cerebellum) and region-specific language areas (BA and WA, as well as larger more inclusive expressive and receptive language regions) for quantification of within-subject variability of activated language clusters. In that paper, Within-Subject Lateralization Index Variability (LIVAR) was calculated across the right-handed patient population to quantify the within-subject variability of holohemispheric and regional language lateralization. LIVAR was defined as (LI1-LI2)/(LI1+LI2) where LI1 and LI2 are the LIs of two consecutive runs within the same scan session. High intrasubject language lateralization repeatability was found using both threshold-dependent and threshold-independent approaches for both holohemispheric and regional language lateralization. Except for a few outliers, within-subject variability of LI from run 1 to run 2 in holohemispheric as well as in local regions for each subject was below 1.0. Variability in LI was lower for the SWG task in comparison to the SC task, which confirmed that the SWG task is a better determinant of language lateralization.

In the same paper, Agarwal et al96 obtained the COM of regional language activation areas to localize the centers of activations of the SWG and SC language tasks. Within-subject Center of Mass Variability (COMVAR) was calculated to quantify the variability in COM location from run 1 (ie, COM1) to run 2 (ie, COM2) of each language task within the same scan session using the Euclidean distance between COM1 and COM2. The displacement of COM coordinates from run 1 to run 2 was found to be within 5 mm. The repeatability in localization centers is demonstrated both when centers of mass of activation are considered within key eloquent regions of the brain, such as BA and WA, as well as in larger more inclusive expressive and receptive language regions. Note that COM of regional language activation areas is usually obtained to localize the centers of activations of each language task; however, it does not necessarily mean that it is representative of the true spatial coordinates of neural activity.97

Back to Top | Article Outline


In this review, current standards and reproducibility of presurgical language mapping with task-based fMRI were discussed. Details of methods of image acquisition, monitoring, and analysis are provided. There are national efforts currently underway to standardize language fMRI for clinical applications, and more work needs to be done to accomplish this fully. Recently, the ASFNR provided guidelines on fMRI paradigm algorithms for presurgical language assessment for language lateralization and localization. In the near future, with the wide range of ongoing research, new windows into brain language function and connectivity will open that will result in improved clinical care.

Back to Top | Article Outline


1. Pillai JJ. The evolution of clinical functional imaging during the past 2 decades and its current impact on neurosurgical planning. AJNR Am J Neuroradiol 2010; 31:219–225.
2. Pillai J, Zaca D, Choudhri A. Clinical impact of integrated physiologic brain tumor imaging. Technol Cancer Res Treat 2010; 9:359–380.
3. Petrella JR, Shah LM, Harris KM, et al. Preoperative functional MR imaging localization of language and motor areas: effect on therapeutic decision making in patients with potentially resectable brain tumors. Radiology 2006; 240:793–802.
4. Holodny AI. Preoperative and Postoperative Mapping of Eloquent Regions in the Brain. ASNR; 2004.
5. Fox PT, Raichle ME. Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects. Proc Natl Acad Sci U S A 1986; 83:1140–1144.
6. Ogawa S, Tank DW, Menon R, et al. Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci U S A 1992; 89:5951–5955.
7. Chen JE, Glover GH. Functional magnetic resonance imaging methods. Neuropsychol Rev 2015; 25:289–313.
8. Lowe MJ, Beall EB. Filippi M. Selection of optimal pulse sequences for fMRI. fMRI Techniques and Protocols. New York, NY:Springer New York; 2016. 69–111.
9. Elster AD. Gradient-echo MR imaging: techniques and acronyms. Radiology 1993; 186:1–8.
10. Hargreaves BA. Rapid gradient-echo imaging. J Magn Reson Imaging 2012; 36:1300–1313.
11. Mansfield P. Multi-planar image formation using NMR spin echoes. J Phys C Solid State Phys 1977; 10:L55.
12. Poustchi-Amin M, Mirowitz SA, Brown JJ, et al. Principles and applications of echo-planar imaging: a review for the general radiologist. Radiographics 2001; 21:767–779.
13. Ahn CB, Kim JH, Cho ZH. High-speed spiral-scan echo planar NMR imaging-I. IEEE Trans Med Imaging 1986; 5:2–7.
14. Bornert P, Schomberg H, Aldefeld B, et al. Improvements in spiral MR imaging. MAGMA 1999; 9:29–41.
15. Deshmane A, Gulani V, Griswold MA, et al. Parallel MR imaging. J Magn Reson Imaging 2012; 36:55–72.
16. Pruessmann KP, Weiger M, Scheidegger MB, et al. SENSE: sensitivity encoding for fast MRI. Magn Reson Med 1999; 42:952–962.
17. Block KT, Frahm J. Spiral imaging: a critical appraisal. J Magn Reson Imaging 2005; 21:657–668.
18. Feinberg DA, Setsompop K. Ultra-fast MRI of the human brain with simultaneous multi-slice imaging. J Magn Reson 2013; 229:90–100.
19. Xu J, Moeller S, Auerbach EJ, et al. Evaluation of slice accelerations using multiband echo planar imaging at 3 T. Neuroimage 2013; 83:991–1001.
20. Cauley SF, Polimeni JR, Bhat H, et al. Interslice leakage artifact reduction technique for simultaneous multislice acquisitions. Magn Reson Med 2014; 72:93–102.
21. Yanasak NE, Kelly MJ. MR imaging artifacts and parallel imaging techniques with calibration scanning: a new twist on old problems. Radiographics 2014; 34:532–548.
22. Engstrom M, Ragnehed M, Lundberg P, et al. Paradigm design of sensory-motor and language tests in clinical fMRI. Neurophysiol Clin 2004; 34:267–277.
23. Benjamin CF, Walshaw PD, Hale K, et al. Presurgical language fMRI: mapping of six critical regions. Hum Brain Mapp 2017; 38:4239–4255.
24. Benjamin CFA, Li AX, Blumenfeld H, et al. Presurgical language fMRI: clinical practices and patient outcomes in epilepsy surgical planning. Hum Brain Mapp 2018; 39:2777–2785.
25. Bookheimer S. Pre-surgical language mapping with functional magnetic resonance imaging. Neuropsychol Rev 2007; 17:145–155.
26. Ives-Deliperi VL, Butler JT, Meintjes EM. Functional MRI language mapping in pre-surgical epilepsy patients: findings from a series of patients in the Epilepsy Unit at Mediclinic Constantiaberg. S Afr Med J 2013; 103:563–567.
27. Genetti M, Grouiller F, Vulliemoz S, et al. Noninvasive language mapping in patients with epilepsy or brain tumors. Neurosurgery 2013; 72:555–565. discussion 65.
28. Mahdavi A, Azar R, Shoar MH, et al. Functional MRI in clinical practice: assessment of language and motor for pre-surgical planning. Neuroradiol J 2015; 28:468–473.
29. Partovi S, Konrad F, Karimi S, et al. Effects of covert and overt paradigms in clinical language fMRI. Acad Radiol 2012; 19:518–525.
30. Szaflarski JP, Gloss D, Binder JR, et al. Practice guideline summary: use of fMRI in the presurgical evaluation of patients with epilepsy: report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology 2017; 88:395–402.
31. Ulmer JL, Hacein-Bey L, Mathews VP, et al. Lesion-induced pseudo-dominance at functional magnetic resonance imaging: implications for preoperative assessments. Neurosurgery 2004; 55:569–579. discussion 80-1.
32. Wilson SM, Bautista A, Yen M, et al. Validity and reliability of four language mapping paradigms. Neuroimage Clin 2017; 16:399–408.
33. Binder JR, Swanson SJ, Hammeke TA, et al. A comparison of five fMRI protocols for mapping speech comprehension systems. Epilepsia 2008; 49:1980–1997.
34. Binder JR, Desai RH, Graves WW, et al. Where is the semantic system? A critical review and meta-analysis of 120 functional neuroimaging studies. Cereb Cortex 2009; 19:2767–2796.
35. Tremblay P, Dick AS. Broca and Wernicke are dead, or moving past the classic model of language neurobiology. Brain Lang 2016; 162:60–71.
36. Roux FE, Draper L, Kopke B, et al. Who actually read Exner? Returning to the source of the frontal “writing centre” hypothesis. Cortex 2010; 46:1204–1210.
37. Binder JR. The Wernicke area: modern evidence and a reinterpretation. Neurology 2015; 85:2170–2175.
38. Mills C, McConnell I. The Naming Center, with the report of a case indicating its location in the temporal lobe. J Nerv Ment Dis 1895; 22:1–7.
39. Hamberger MJ, Seidel WT, McKhann GM 2nd, et al. Brain stimulation reveals critical auditory naming cortex. Brain 2005; 128:2742–2749.
40. Gaillard WD, Balsamo LM, Ibrahim Z, et al. fMRI identifies regional specialization of neural networks for reading in young children. Neurology 2003; 60:94–100.
41. Hirsch J, Ruge MI, Kim KH, et al. An integrated functional magnetic resonance imaging procedure for preoperative mapping of cortical areas associated with tactile, motor, language, and visual functions. Neurosurgery 2000; 47:711–721. discussion 721-722.
42. Pillai JJ, Zaca D. Relative utility for hemispheric lateralization of different clinical fMRI activation tasks within a comprehensive language paradigm battery in brain tumor patients as assessed by both threshold-dependent and threshold-independent analysis methods. Neuroimage 2011; 54 (Suppl 1):S136–S145.
43. FitzGerald DB, Cosgrove GR, Ronner S, et al. Location of language in the cortex: a comparison between functional MR imaging and electrocortical stimulation. AJNR Am J Neuroradiol 1997; 18:1529–1539.
44. Zaca D, Nickerson JP, Deib G, et al. Effectiveness of four different clinical fMRI paradigms for preoperative regional determination of language lateralization in patients with brain tumors. Neuroradiology 2012; 54:1015–1025.
45. Yetkin FZ, Swanson S, Fischer M, et al. Functional MR of frontal lobe activation: comparison with Wada language results. AJNR Am J Neuroradiol 1998; 19:1095–1098.
46. Pillai JJ, Allison JD, Sethuraman S, et al. Functional MR imaging study of language-related differences in bilingual cerebellar activation. AJNR Am J Neuroradiol 2004; 25:523–532.
47. Pillai JJ, Araque JM, Allison JD, et al. Functional MRI study of semantic and phonological language processing in bilingual subjects: preliminary findings. Neuroimage 2003; 19:565–576.
48. Laurito JT, Bryan RN, Mathews UP, Ulmer JU, Lowe MJ. Functional Brain Mapping, Categorical Course in Diagnostic Radiology: Neuroradiology. Oak Brook, IL: Radiological Society of North America RSNA; 2000.
49. Salvan CV, Ulmer JL, DeYoe EA, et al. Visual object agnosia and pure word alexia: correlation of functional magnetic resonance imaging and lesion localization. J Comput Assist Tomogr 2004; 28:63–67.
50. Phillips MD, Lowe MJ, Lurito JT, et al. Temporal lobe activation demonstrates sex-based differences during passive listening. Radiology 2001; 220:202–207.
51. Zaca D, Jarso S, Pillai JJ. Role of semantic paradigms for optimization of language mapping in clinical FMRI studies. AJNR Am J Neuroradiol 2013; 34:1966–1971.
52. Heeger DJ, Ress D. What does fMRI tell us about neuronal activity? Nat Rev Neurosci 2002; 3:142–151.
53. Vanzetta I, Grinvald A. Increased cortical oxidative metabolism due to sensory stimulation: implications for functional brain imaging. Science 1999; 286:1555–1558.
54. Detre JA, Wang J. Technical aspects and utility of fMRI using BOLD and ASL. Clin Neurophysiol 2002; 113:621–634.
55. Fox PT, Raichle ME, Mintun MA, et al. Nonoxidative glucose consumption during focal physiologic neural activity. Science 1988; 241:462–464.
56. Broca P. Remarks on the Seat of the Faculty of Articulated Language, Following an Observation of Aphemia (Loss of Speech). Surgeon of the Bicêtre Hosptial. First published in Bulletin de la Société Anatomique 1861;6, 330–357.
57. Wada JA, Rasmussen T. Intracarotid injection of sodium amytal for the lateralization of cerebral speech dominance. J Neurosurg 1960; 17:266–282.
58. Binder JR, Swanson SJ, Hammeke TA, et al. Determination of language dominance using functional MRI: a comparison with the Wada test. Neurology 1996; 46:978–984.
59. Rutten GJ, Ramsey NF, van Rijen PC, et al. FMRI-determined language lateralization in patients with unilateral or mixed language dominance according to the Wada test. Neuroimage 2002; 17:447–460.
60. Hertz-Pannier L, Gaillard WD, Mott SH, et al. Noninvasive assessment of language dominance in children and adolescents with functional MRI: a preliminary study. Neurology 1997; 48:1003–1012.
61. Benson RR, FitzGerald DB, LeSueur LL, et al. Language dominance determined by whole brain functional MRI in patients with brain lesions. Neurology 1999; 52:798–809.
62. Hund-Georgiadis M, Lex U, Friederici AD, et al. Non-invasive regime for language lateralization in right- and left-handers by means of functional MRI and dichotic listening. Exp. Brain Res 2002; 145:166–176.
63. Binder JR. Functional MRI is a valid noninvasive alternative to Wada testing. Epilepsy Behav 2011; 20:214–222.
64. Bradshaw AR, Bishop DVM, Woodhead ZVJ. Methodological considerations in assessment of language lateralisation with fMRI: a systematic review. PeerJ 2017; 5:e3557.
65. Springer JA, Binder JR, Hammeke TA, et al. Language dominance in neurologically normal and epilepsy subjects: a functional MRI study. Brain 1999; 122:2033–2046.
66. Adcock JE, Wise RG, Oxbury JM, et al. Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage 2003; 18:423–438.
67. Branco DM, Suarez RO, Whalen S, et al. Functional MRI of memory in the hippocampus: laterality indices may be more meaningful if calculated from whole voxel distributions. Neuroimage 2006; 32:592–602.
68. Suarez RO, Whalen S, Nelson AP, et al. Threshold-independent functional MRI determination of language dominance: a validation study against clinical gold standards. Epilepsy Behav 2009; 16:288–297.
69. Gaillard WD, Balsamo L, Xu B, et al. Language dominance in partial epilepsy patients identified with an fMRI reading task. Neurology 2002; 59:256–265.
70. Ojemann GA. Individual variability in cortical localization of language. J Neurosurg 1979; 50:164–169.
71. Gartus A, Foki T, Geissler A, et al. Improvement of clinical language localization with an overt semantic and syntactic language functional MR imaging paradigm. AJNR Am J Neuroradiol 2009; 30:1977–1985.
72. Bahn MM, Lin W, Silbergeld DL, et al. Localization of language cortices by functional MR imaging compared with intracarotid amobarbital hemispheric sedation. AJR Am J Roentgenol 1997; 169:575–579.
73. Bizzi A, Blasi V, Falini A, et al. Presurgical functional MR imaging of language and motor functions: validation with intraoperative electrocortical mapping. Radiology 2008; 248:579–589.
74. Giussani C, Roux FE, Ojemann J, et al. Is preoperative functional magnetic resonance imaging reliable for language areas mapping in brain tumor surgery? Review of language functional magnetic resonance imaging and direct cortical stimulation correlation studies. Neurosurgery 2010; 66:113–120.
75. Kuchcinski G, Mellerio C, Pallud J, et al. Three-tesla functional MR language mapping: comparison with direct cortical stimulation in gliomas. Neurology 2015; 84:560–568.
76. Meier MP, Ilmberger J, Fesl G, et al. Validation of functional motor and language MRI with direct cortical stimulation. Acta Neurochir (Wien) 2013; 155:675–683.
77. Rutten GJ, Ramsey NF, van Rijen PC, et al. Development of a functional magnetic resonance imaging protocol for intraoperative localization of critical temporoparietal language areas. Ann Neurol 2002; 51:350–360.
78. Yetkin FZ, Mueller WM, Morris GL, et al. Functional MR activation correlated with intraoperative cortical mapping. AJNR Am J Neuroradiol 1997; 18:1311–1315.
79. Lurito JT, Lowe MJ, Sartorius C, et al. Comparison of fMRI and intraoperative direct cortical stimulation in localization of receptive language areas. J Comput Assist Tomogr 2000; 24:99–105.
80. Tomczak RJ, Wunderlich AP, Wang Y, et al. fMRI for preoperative neurosurgical mapping of motor cortex and language in a clinical setting. J Comput Assist Tomogr 2000; 24:927–934.
81. Pouratian N, Bookheimer SY, Rex DE, et al. Utility of preoperative functional magnetic resonance imaging for identifying language cortices in patients with vascular malformations. Neurosurg Focus 2002; 13:e4.
82. Signorelli F, Guyotat J, Schneider F, et al. Technical refinements for validating functional MRI-based neuronavigation data by electrical stimulation during cortical language mapping. Minim Invasive Neurosurg 2003; 46:265–268.
83. Roux FE, Boulanouar K, Lotterie JA, et al. Language functional magnetic resonance imaging in preoperative assessment of language areas: correlation with direct cortical stimulation. Neurosurgery 2003; 52:1335–1345. discussion 45–47.
84. Voyvodic JT. Reproducibility of single-subject fMRI language mapping with AMPLE normalization. J Magn Reson Imaging 2012; 36:569–580.
85. Poline JB, Strother SC, Dehaene-Lambertz G, et al. Motivation and synthesis of the FIAC experiment: reproducibility of fMRI results across expert analyses. Hum Brain Mapp 2006; 27:351–359.
86. Morrison MA, Churchill NW, Cusimano MD, et al. Reliability of task-based fMRI for preoperative planning: a test-retest study in brain tumor patients and healthy controls. PLoS One 2016; 11:e0149547.
87. Harrington GS, Buonocore MH, Farias ST. Intrasubject reproducibility of functional MR imaging activation in language tasks. AJNR Am J Neuroradiol 2006; 27:938–944.
88. Carp J. Better living through transparency: improving the reproducibility of fMRI results through comprehensive methods reporting. Cogn Affect Behav Neurosci 2013; 13:660–666.
89. Fernandez G, Specht K, Weis S, et al. Intrasubject reproducibility of presurgical language lateralization and mapping using fMRI. Neurology 2003; 60:969–975.
90. Weng HH, Noll KR, Johnson JM, et al. Accuracy of presurgical functional MR imaging for language mapping of brain tumors: a systematic review and meta-analysis. Radiology 2018; 286:512–523.
91. Rombouts SA, Barkhof F, Hoogenraad FG, et al. Test-retest analysis with functional MR of the activated area in the human visual cortex. AJNR Am J Neuroradiol 1997; 18:1317–1322.
92. Stevens MT, Clarke DB, Stroink G, et al. Improving fMRI reliability in presurgical mapping for brain tumours. J Neurol Neurosurg Psychiatry 2016; 87:267–274.
93. Maldjian JA, Laurienti PJ, Driskill L, et al. Multiple reproducibility indices for evaluation of cognitive functional MR imaging paradigms. AJNR Am J Neuroradiol 2002; 23:1030–1037.
94. Caceres A, Hall DL, Zelaya FO, et al. Measuring fMRI reliability with the intra-class correlation coefficient. Neuroimage 2009; 45:758–768.
95. Black DF, Vachha B, Mian A, et al. American Society of Functional Neuroradiology-recommended fMRI paradigm algorithms for presurgical language assessment. AJNR Am J Neuroradiol 2017; 38:E65–E73.
96. Agarwal S, Hua J, Sair HI, et al. Repeatability of language fMRI lateralization and localization metrics in brain tumor patients. Hum Brain Mapp 2018; 39:4733–4742.
97. Fesl G, Braun B, Rau S, et al. Is the center of mass (COM) a reliable parameter for the localization of brain function in fMRI? Eur Radiol 2008; 18:1031–1037.

language fMRI; lateralization; presurgical mapping; repeatability; reproducibility

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.