Atormac
Neurology India
Open access journal indexed with Index Medicus
  Users online: 2002  
 Home | Login 
  About Current Issue Archive Ahead of print Search Instructions Online Submission Subscribe Etcetera Contact  
  Navigate Here 
 Search
 
  
 Resource Links
  »  Similar in PUBMED
 »  Search Pubmed for
 »  Search in Google Scholar for
 »Related articles
  »  Article in PDF (875 KB)
  »  Citation Manager
  »  Access Statistics
  »  Reader Comments
  »  Email Alert *
  »  Add to My List *
* Registration required (free)  

 
  In this Article
 »  Abstract
 »  Localization of ...
 »  Presurgical Eloq...
 » Conclusion
 »  References
 »  Article Figures
 »  Article Tables

 Article Access Statistics
    Viewed814    
    Printed34    
    Emailed0    
    PDF Downloaded55    
    Comments [Add]    

Recommend this journal

 


 
Table of Contents    
REVIEW ARTICLE
Year : 2017  |  Volume : 65  |  Issue : 7  |  Page : 25-33

Use of resting-state fMRI in planning epilepsy surgery


1 Medical Student MSIII, Baylor College of Medicine; Department of Statistics, Rice University, Houston, Texas, USA
2 Department of Neurology, Baylor College of Medicine; Neurology Care Line, Micheal E DeBakey VA Medical Center, Houston, Texas, USA
3 Department of Neurology, University of California, Los Angeles, California, USA
4 Departments of Neurology; Neurobiology; Psychiatry and Biobehavioral Sciences; The Brain Research Institute, University of California, Los Angeles, California, USA

Date of Web Publication8-Mar-2017

Correspondence Address:
Sharon Chiang
Baylor College of Medicine, Houston, Texas
USA
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/neuroindia.NI_823_16

Rights and Permissions

 » Abstract 

Epileptic seizures result from abnormal neuronal excitability and synchronization, affecting 0.5–1% of the population worldwide. Although anti-seizure drugs are often effective, a significant number of patients with epilepsy continue to experience refractory seizures and are candidates for surgical resection. Whereas standard presurgical evaluation has relied on intracranial electroencephalography (icEEG) and direct cortical stimulation to identify epileptogenic tissue and areas of cortex for which resection would produce clinical deficits, the invasive nature and limited spatial extent of icEEG has led to the investigation of less invasive imaging modalities as adjunctive tools in the presurgical workup. In the past few decades, functional connectivity MRI has emerged as a promising approach for presurgical mapping, leading to a surge in the number of proposed methods and biomarkers for identifying epileptogenic tissue. This review focuses on recent advances in the use of functional connectivity MRI toward its application for presurgical planning, including epilepsy localization and eloquent cortex mapping.


Keywords: Eloquent cortex, epilepsy, epileptogenic zone, functional connectivity magnetic resonance imaging, presurgical planning
Key Messages:

  • Resting-state functional MRI has shown potential as an additional tool for non-invasive presurgical evaluation in patients with drug resistant epilepsy.
  • New developments in multimodal image analysis and dynamic functional connectivity are anticipated to further the ability of resting-state functional MRI in identifying markers for epileptogenicity and mapping eloquent cortex.


How to cite this article:
Chiang S, Haneef Z, Stern JM, Engel J. Use of resting-state fMRI in planning epilepsy surgery. Neurol India 2017;65, Suppl S1:25-33

How to cite this URL:
Chiang S, Haneef Z, Stern JM, Engel J. Use of resting-state fMRI in planning epilepsy surgery. Neurol India [serial online] 2017 [cited 2017 Nov 22];65, Suppl S1:25-33. Available from: http://www.neurologyindia.com/text.asp?2017/65/7/25/201675


Epilepsy is one of the most prevalent, serious, chronic neurological disorders, affecting nearly 0.5–1% of the population worldwide.[1] For the 20–30% of patients suffering from epilepsy whose seizures are resistant to medications,[2] surgical resection of epileptogenic tissue may be an effective method of treatment. However, determining candidacy for epilepsy surgery requires sophisticated testing to precisely delineate the epileptogenic tissue and its anatomic relationship with the eloquent cortex. This is needed to achieve postoperative seizure freedom while minimizing the likelihood of postoperative focal neurologic deficits.

Current standard noninvasive presurgical evaluation employs multiple imaging and electrophysiological modalities, almost always including structural magnetic resonance imaging (MRI) and surface electroencephalography (EEG). However, such noninvasive assessment is often limited by normal structural MRI and nonfocal surface EEG. As such, it is nonconfirmatory for the epileptogenic zone (EZ) in a considerable proportion of patients with pharmacoresistant epilepsy.[3],[4] Intracranial EEG (icEEG) remains the necessary test in such situations for delineation of the EZ, albeit with limitations, including the extent of coverage possible. As the invasive nature of intracranial EEG carries risks, noninvasive methods are preferable.

The goal of the evaluation is the identification of the EZ, which is defined as the area necessary and sufficient for spontaneous seizure generation. Recent studies, however, have broadened this concept, suggesting that ictal activity may begin in a spatially localized EZ which spreads to, and is modulated by, a larger network of brain structures.[5] In contrast to the spatially localized EZ, an epileptogenic network is dynamic and may change with a longer epilepsy duration.[6] Strategic disruption of the epileptogenic network may allow for seizure elimination while decreasing the risk of recurrence. This approach is now being conceptualized as results from investigations into the epileptic network are being reported. If effective, the epileptic network characterization would alter the approach to epilepsy surgery.

There has been an increasing interest in functional MRI (fMRI) investigations of the human connectome, and such investigations have considerable value in understanding epilepsy networks with a potential to advance surgical treatment. This use of fMRI differs from the more established approach at many epilepsy surgery centers of presurgically mapping the eloquent cortex, and sometimes includes a simultaneous EEG to localize regions generating interictal or ictal activity.[7],[8],[9],[10] For mapping the eloquent function or epileptic activity, the approach seeks to identify areas of cortex with fMRI blood-oxygen-level dependent (BOLD) signal activations associated with particular tasks. such as in eloquent cortex mapping, or in response to EEG events, as in the identification of interictal or ictal-onset zones. This use of fMRI has at times supplanted the traditional approaches for identifying the eloquent cortex, such as the Wada test for language lateralization, or cortical mapping by electrical stimulation during electrocorticography. As Wada testing may be used to lateralize but not localize language, and because of the invasive nature of direct cortical stimulation, task-based fMRI can be superior in both anatomic detail and scope, as well as safer by virtue of its noninvasiveness. Questions relating to the optimal task choice, as well as the extrapolation of task designs from healthy to patient populations, however, provide a challenge. As task-based fMRI depends on the patient's ability to perform the needed task, it may be suboptimal in children and in patients with neurological deficits, altered mental status, or under sedation. In addition, the need to perform each task separately requires long acquisition times if multiple functional sites are to be investigated.

Functional connectivity MRI (fcMRI) differs from task-based fMRI by recording a resting-state. Rather than performing an explicit task, the patient is asked to lie still and not to think about anything in particular. It is during this undirected mental state that the default mode network is active.[11] Instead of associating the fMRI BOLD signal to a behavior or EEG finding, connectivity is calculated from spontaneous BOLD signal fluctuations and identifying the relationship in the fluctuations across cerebral regions. Estimated correlations between low-frequency fluctuations (<0.1 Hz) have been found to reproducibly reflect the underlying intrinsic neural architecture,[12] and are presumed to be generated by underlying fluctuations in neural activity. These correlations may be complementary in identifying seizure-generating networks as well as defining eloquent cortex or networks. Graph theory provides an additional network-level perspective by quantifying how each given voxel or region-of-interest is connected to the remainder of the network. Network “hubs,” or regions, which are central to the functioning of the network, may be quantified through measures such as degree or betweenness centrality, whereas measures of local and/or long-range connectivity may be quantified through measures such as the clustering coefficient or path length, respectively.[13]

With respect to both localization of the EZ and eloquent cortex mapping, there is a need for less invasive and more reliable diagnostic tools. In this article, we review the converging evidence from fMRI connectomic analyses toward their application to presurgical planning. In particular, we focus on the clinical impact of this evidence for localization of epileptogenic tissue and eloquent cortex mapping. A summary of the evidence underlying the following discussion is provided in [Table 1] and [Table 2].
Table 1: Summary of clinical studies on fcMRI in presurgical localization

Click here to view
Table 2: Summary of clinical studies on fcMRI in presurgical eloquent cortex mapping

Click here to view



 » Localization of Epileptogenic Tissue Top


Electroencephalography–functional magnetic resonance imaging

Interictal epileptiform discharges (IEDs) reflect abnormal interictal neural activity and arise from the so-called irritative zone (IZ). One approach for mapping the IZ utilizes simultaneous EEG–fMRI recordings, which combine the advantages of the specificity of EEG for epileptic abnormality and the spatial resolution of fMRI to map changes in the cerebral hemodynamic flow corresponding to IEDs. IEDs are identified in EEG data, convolved with a hemodynamic response function (HRF), and treated as regressors of interest in a General Linear Model (GLM). Areas that are activated during IEDs are then identified as the IZ [Figure 1]. While EEG has classically been used for IZ identification, the spatial resolution and tomography of fMRI is significantly higher than EEG, permitting a depiction of the IZ's hemodynamic map. Studies report that greater correspondence between the IZ delineated by EEG–fMRI and resected tissue leads to a greater probability of postoperative seizure freedom. One group found that 58% of the patients with resection areas concordant with the IZ and 19% of the patients with discordant areas attain good postsurgical seizure outcomes (Engel 1-2).[14] Similarly, 80% of the patients with concordant IZ and seizure-onset zone (SOZ) attained >50% reduction in seizure frequency one year after surgery, compared to only 17% of the patients with discordant IZ and SOZ.[15] Such studies suggest that simultaneous EEG–fMRI may be a useful adjunct in the preoperative evaluation of epileptogenic cortex.
Figure 1: Simultaneous EEG–fMRI. In simultaneous EEG–fMRI, EEG and fMRI are acquired simultaneously. Interictal epileptiform discharges (red waveform) may then be convolved with a hemodynamic response function (HRF) and utilized within the General Linear Model (GLM) to identify areas with BOLD signal activation (red area) in response to IEDs

Click here to view



However, the EEG–fMRI approach to localizing epileptogenic tissue suffers from a few limitations:

  • The IZ defined by IED-associated BOLD maps includes all regions experiencing BOLD increases concurrent with IEDs, regardless of the level of irritability. Areas within the mapped IZ, however, may have different likelihoods of irritability and probabilities of postsurgical seizure freedom, if resected. Effective connectivity analysis of areas implicated in the IZ, which seeks to identify the directionality of connectivity, may provide a way to identify “driving areas” within the IZ, which are more important for resection. In one case report, failure to resect the presurgically identified “driving area” was found to result in postsurgical seizure recurrence.[16] However, validation in larger samples is needed to establish the clinical utility of this approach. New developments in statistical methodology may improve the accuracy of effective connectivity inference by incorporating external information from DTI or MRI, as well as statistical borrowing of information across patients to improve inference on the single-patient level.[17],[18],[19] Other issues with the EEG–fMRI approach stem from inherent assumptions within the approach, several of which are listed below:
  • Explicit specification of the HRF is required. However, the HRF has been found to vary across patients, recording sessions, and brain regions, which may be related to abnormality within the brain tissue.[20],[21],[22] Epilepsy may cause variation of the HRF,[23] leading to misspecification and poor downstream statistical inference,
  • A significant proportion of patients with epilepsy may not exhibit IEDs during the brief scalp EEG recordings obtained during fMRI, with many patients exhibiting either normal EEGs or nonspecific slowing,[24]
  • Scalp EEG used during EEG–fMRI has limited sensitivity to small IEDs and IEDs that do not project to the scalp.[25] Newer studies using icEEG during EEG–fMRI will be helpful in eliminating this limitation, but are more invasive
  • Clustering of IED events may result in violations of linearity assumptions for neurovascular coupling.[26]


Functional connectivity magnetic resonance imaging

fcMRI provides a promising alternative approach for identifying the IZ without necessitating assumptions of the GLM. In particular, IED-associated changes in the graph theory measures of network properties, rather than in BOLD amplitude, may facilitate IZ delineation. As graph measures are calculated based on the number of connections for each voxel, this approach obviates the need for GLM assumptions. One approach, for example, spatially maps the IZ using IED-associated changes in local degree centrality, a graph measure of local network connectivity. The clinically presumed EZ had increases in local degree centrality during IEDs compared to baseline. Furthermore, whereas IZ identified using this method was concordant with the presumed EZ in all six patients in the sample, GLM-based methods were concordant in only three of the six patients.[26] Although large sample validation is needed, the potential promise of non-task-based approaches to IZ mapping may be because of the ability to detect BOLD changes not identifiable under GLM-based assumptions, such as when IEDs do not appear on the scalp EEG or when IEDs cluster in a short period, violating assumptions of linear neurovascular coupling.[26] Various fcMRI graph theory studies have supported the potential for resting-state graph theory measures to lateralize the clinically presumed epileptogenic focus.[27],[28],[29],[30] However, fewer studies have addressed the sensitivity and specificity of these measures for focus lateralization.[31],[32],[33]

Functional connectivity studies have also shown that abnormalities in the strength of functional connectivity in the absence of IED information may serve as a marker for the EZ. The direction in which functional connectivity changes near the suspected EZ has not been definitely established. While several fcMRI studies have also found evidence that local functional connectivity is increased for epileptogenic tissue,[34],[35],[36] other research has shown decreases.[33],[37] Indeed, icEEG studies have shown increases in functional connectivity in the EZ.[38],[39] Discrepant conclusions may potentially be due to patient heterogeneity, small sample sizes, varying definitions of local neighborhoods, and different local connectivity metrics and modalities. Differences between the SOZ,[35],[37] IZ,[34] and a priori defined regions of interest (ROIs)[33],[36] may also contribute. Recent evidence furthermore indicates that connectivity is dynamic both within and across scans,[40],[41],[42] such that traditional static measures of connectivity provide only an average measure of functional connectivity. Local measures of connectivity exhibit significant temporal nonstationarity,[42] potentially impacting whether increases or decreases in functional connectivity (FC) are identified. Discrepant results from fcMRI compared to icEEG may also result from neurovascular decoupling in epilepsy and differences in timescales.[33],[39]

Other changes in brain connectivity in epilepsy

Other changes have also been identified in brain connectivity that contribute to understanding the pathophysiology of epilepsy and may provide future targets for intervention by surgery or devices. Connectivity to spatially remote regions appears altered for the presumed EZ and may indicate diffuse epileptic networks. However, the direction of this alteration is not consistent. Some studies have investigated FC between the IZ and remote regions and found evidence that FC is decreased,[34] whereas other studies on the SOZ have found that connectivity to remote regions is increased. Among patients with left TLE, the superior left temporal pole has been found to have increased connectivity with neighboring regions such as within the left temporal and frontal cortex; however, decreased connectivity to more remote regions such as bilateral dorsal posterior cingulate cortex, contralateral inferior temporal gyrus, and contralateral thalamus occurs.[36] Another study of patients with TLE found that the hippocampus has increased connectivity to key areas of the limbic system and frontal lobes, and decreased connectivity to various sensorimotor areas.[30] These variations may reflect the fact that increases in remote functional connectivity may be less specific to epileptogenic tissue than increases in local functional connectivity.[35] In general, data from EEG–fMRI suggests that more diffuse epileptogenic networks are associated with poor postoperative outcomes.[14],[15],[43] Measures of the diffusivity of epileptogenic networks, such as laterality indices of functional connectivity computed from the presumed epileptogenic focus,[43] may be useful for the prediction of surgical success.

Some studies have investigated whether combining local and remote connectivity may be useful as a marker for localization,[37],[44] based on the hypothesis that contralateral changes in connectivity may result as a compensatory mechanism for abnormalities ipsilateral to the seizure focus. As various aspects of functional connectivity appear to be affected by the progression of epilepsy, including homogenization of connectivity strengths [45] and the diminishment of functional connectivity in ipsilateral networks,[46] markers which account for both local and remote changes may be useful in providing markers less dependent on epilepsy burden. Various approaches to considering relative amounts of ipsilateral and contralateral connectivity have been considered, including computing voxel-wise difference maps between intra- and interhemispheric connectivity,[44] ROI-based approaches which compare the local connectivity in an ROI containing the SOZ to that contralateral to the clinically presumed epileptogenic focus,[37] and ROI-based approaches which compare ipsilateral versus contralateral connectivity among predefined ROIs thought to be typically involved in epileptogenic networks.[33] These studies consistently suggest that functional connectivity is decreased ipsilateral to the seizure focus relative to the contralateral hemisphere. Less chaotic BOLD dynamics within the functional deficit zone, defined as the area of cortex that is interictally dysfunctional, have also been identified in association with increased local functional connectivity, which may reflect increased density of inhibitory or excitatory inputs.[36]


 » Presurgical Eloquent Cortex Mapping Top


Task-based fMRI has traditionally been used for the preoperative delineation of eloquent cortex. During a typical task-based fMRI session, a stimulus is administered to the patient. Finger-tapping paradigms are commonly used to localize sensorimotor cortex, whereas verb-generation or sentence-completion paradigms are commonly used to localize language cortex. Areas with BOLD signal change associated with these tasks are identified as the corresponding cortical representation area of the cortex. Resting-state fMRI (rs-fMRI) provides a noninvasive approach and allows for the simultaneous mapping of multiple functions. Several approaches have been proposed for applying rs-fMRI for preoperative localization of eloquent cortex. This would obviate the task paradigm and would be easier for the patient and potentially be more reliable. The resting-state approach has been directed toward sensorimotor and language networks in particular.

A traditional approach to mapping functional networks involves seed-based cross-correlation analysis [Figure 2]a. In this approach, a seed voxel or cluster is determined based on brain anatomy or previous functional studies, and the functional network is then extracted by computing the univariate correlations between the BOLD signal of each voxel with the seed region.[47] Seed-based cross-correlation analysis has been found to yield sensorimotor maps consistent with electrical stimulation mapping (ESM), and to perform at least as well as task-based fMRI for consistency with ESM.[48],[49] An alternative data-driven technique for identifying functional networks in rs-fMRI data, which does not require predefined seeds, is spatial independent component analysis (ICA). In spatial ICA, the observed rs-fMRI data is mathematically decomposed into spatially nonoverlapping and temporally coherent regions, which correspond to various resting-state networks [Figure 2]b.[50],[51] Several studies have consistently shown that ICA can identify sensorimotor networks in healthy individuals.[52],[53],[54],[55],[56] Studies in patients with epilepsy or brain tumors have also found that ICA of rs-fMRI data produces sensorimotor maps similar to those obtained from task-based GLM mapping.[57] Comparisons of maps elicited through rs-fMRI also appear to be consistent with sensorimotor maps elicited through ESM in patients with epilepsy.[49],[58] In general, seed-based approaches have been found to be potentially problematic and generally inferior to ICA because of the sensitivity of seed placement.[59] ICA, which does not require predefined seeds, is thought to offer an advantage in this case. However, ICA has its own drawbacks, including the need to specify the number of independent components as well as for user intervention to identify which components reflect neural systems versus artifacts. Template matching methods have been proposed in which group-level ICA is used to identify appropriate networks on the individual patient level.[60] Other approaches avoid reliance on templates by combining patient-level data into a single ICA followed by back-reconstruction of individual patient ICs,[61] or other approaches such as “partner-matching,” which identifies reproducible components by clustering components within or across patients.[61] In addition, findings from ICA can also be influenced by the number of ICs if the number of ICs is too small.[59] A critical number of ICs below which ICA results are less robust was estimated empirically as nine,[59] although many studies choose a relatively higher order ICA (>60 ICs),[51],[57],[62] which reduces the loss of important information within a reasonable computation time. In the case of the sensorimotor cortex, seed-based mapping may efficiently map sensorimotor cortex as indicated by comparisons with ESM, and appears to yield similar maps to those obtained through ICA.
Figure 2: Seed-based cross-correlation analysis and spatial independent component analysis (ICA) may be used to map functional networks. (a) In seed-based analyses, the average BOLD time-series in a predefined seed ROI (shown in red) is correlated with the time-courses of all other voxels in the brain (representative black arrows shown). A threshold is then applied to identify voxels significantly correlated with the seed ROI. (b) In spatial ICA, the fMRI data is decomposed into a “source matrix,” which yields spatially nonoverlapping and temporally coherent regions, and a “mixing matrix,” which yields the time-series of each source

Click here to view


Identification of the language area is more difficult because of the greater locational variability across individuals. Furthermore, patients with epilepsy may have additional lesion-induced architectural rearrangements, further increasing the variability of the language network's spatial distribution.[63] Therefore, employing seed-based cross-correlation analysis is more challenging for the language network, and particularly for patients with abnormal brain anatomy. ICA or neural network methods may hold more promise than seed-based methods for mapping language cortex. One recent study on healthy controls has shown that blind-source separation of rs-fMRI data using ICA, followed by semi-automatic classification of the language network, yields good overlap with the language network obtained from task fMRI.[64] Recent studies to evaluate the feasibility of rs-fMRI for identifying language networks among patients with epilepsy have also yielded favorable results. An initial study among six patients with temporal and frontal lobe epilepsies showed fair sensitivity and specificity for rs-fMRI language mapping compared to ESM when back-propagation in artificial neural networks was used.[58] While promising, the reliance of this approach on an adequate training set may lead to difficulties, particularly when dealing with patients whose brains exhibit developmental or plastic effects. Future developments in data-sharing collaborations, such as through the Human Connectome Project [65] and the Brain Genomics Superstruct Project,[66] may be helpful in this regard. ICA provides an avenue for extracting functional networks while avoiding reliance on a training set. A fair amount of overlap has been found between language networks extracted using ICA from rs-fMRI data compared to the language network obtained from task fMRI, in a mixed group of 12 patients with brain tumors and three patients with epilepsy. Furthermore, stability of language IC maps was found to be relatively stable above 50 components.[67]


 » Conclusion Top


The attainment of seizure freedom while preserving functions for patients with medication-resistant epilepsy depends on the precision with which areas with the greatest likelihood to compose the epileptogenic zone can be identified. While icEEG remains the gold standard for identifying the epileptogenic zone, the spatial coverage and noninvasive nature of rs-fMRI suggests its potential as an adjunct in presurgical EZ localization and eloquent cortex mapping. Rapidly evolving research involving multimodal imaging and dynamic functional connectivity is anticipated to further the ability of rs-fMRI to identify markers for epileptogenicity. These developments may allow for better understanding of how epileptogenesis results from abnormal connectivity, as well as guide more effective surgical interventions.

Acknowledgements

Funding/support for this research was provided by: (1) NIH-NINDS K23 Grant NS044936 (JMS); (2) The Leff Family Foundation (JMS); (3) P01 NS02808 (JE); (4) R01 NS33310 (JE); (5) U01 NS42372 (JE); (6) P20 NS80181 (JE).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest

 
 » References Top

1.
Sander JW. The epidemiology of epilepsy revisited. Curr Opin Neurol 2003;16:165-70.  Back to cited text no. 1
    
2.
Schuele SU, Luders HO. Intractable epilepsy: Management and therapeutic alternatives. Lancet Neurol 2008;7:514-24.  Back to cited text no. 2
    
3.
Berg AT, Vickrey BG, Langfitt JT, Sperling MR, Walczak TS, Shinnar S, Bazil CW, Pacia SV, Spencer SS. The multicenter study of epilepsy surgery: Recruitment and selection for surgery. Epilepsia 2003;44:1425-33.  Back to cited text no. 3
    
4.
Spencer SS. Substrates of localization-related epilepsies: Biologic implications of localizing findings in humans. Epilepsia 1998;39:114-23.  Back to cited text no. 4
    
5.
Engel J, Jr., Thompson PM, Stern JM, Staba RJ, Bragin A, Mody I. Connectomics and epilepsy. Current Opinion in Neurology 2013;26:186-94.  Back to cited text no. 5
    
6.
van Dellen E, Douw L, Baayen JC, Heimans JJ, Ponten SC, Vandertop WP, Velis DN, Stam CJ, Reijneveld JC. Long-term effects of temporal lobe epilepsy on local neural networks: A graph theoretical analysis of corticography recordings. PLoS One 2009;4(11): e8081.  Back to cited text no. 6
    
7.
Detre JA. fMRI: Applications in epilepsy. Epilepsia 2004;45 Suppl 4: 26-31.  Back to cited text no. 7
    
8.
Gaillard WD, Balsamo L, Xu B, McKinney C, Papero PH, Weinstein S, et al. fMRI language task panel improves determination of language dominance. Neurology 2004;63:1403-8.  Back to cited text no. 8
    
9.
Laufs H, Duncan JS. Electroencephalography/functional MRI in human epilepsy: What it currently can and cannot do. Curr Opin Neurol 2007;20:417-23.  Back to cited text no. 9
    
10.
Zijlmans M, Huiskamp G, Hersevoort M, Seppenwoolde JH, van Huffelen AC, Leijten FS. EEG-fMRI in the preoperative work-up for epilepsy surgery. Brain 2007;130(Pt 9):2343-53.  Back to cited text no. 10
    
11.
Buckner RL, Andrews-Hanna JR, Schacter DL. The brain's default network: Anatomy, function, and relevance to disease. Ann N Y Acad Sci 2008;1124:1-38.  Back to cited text no. 11
    
12.
Biswal B, Zerrin Yetkin F, Haughton VM, Hyde JS. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med 1995;34:537-541.  Back to cited text no. 12
    
13.
Chiang S, Haneef Z. Graph theory findings in the pathophysiology of temporal lobe epilepsy. Clin Neurophysiol 2014;125:1295-305.  Back to cited text no. 13
    
14.
An D, Fahoum F, Hall J, Olivier A, Gotman J, Dubeau F. Electroencephalography/functional magnetic resonance imaging responses help predict surgical outcome in focal epilepsy. Epilepsia 2013;54:2184-94.  Back to cited text no. 14
    
15.
Thornton R, Vulliemoz S, Rodionov R, Carmichael DW, Chaudhary UJ, Diehl B, et al. Epileptic networks in focal cortical dysplasia revealed using electroencephalography-functional magnetic resonance imaging. Ann Neurol 2011;70:822-37.  Back to cited text no. 15
    
16.
Vaudano AE, Avanzini P, Tassi L, Ruggieri A, Cantalupo G, Benuzzi F, et al. Causality within the epileptic network: An EEG-fMRI study validated by intracranial EEG. Front Neurol 2013;4:185.  Back to cited text no. 16
    
17.
Gorrostieta C, Fiecas M, Ombao H, Burke E, Cramer S. Hierarchical vector auto-regressive models and their applications to multi-subject effective connectivity. Front Comput Neurosci 2013;7:159.  Back to cited text no. 17
    
18.
Stephan KE, Tittgemeyer M, Knosche TR, Moran RJ, Friston KJ. Tractography-based priors for dynamic causal models. Neuroimage 2009;47:1628-38.  Back to cited text no. 18
    
19.
Chiang S, Guindani M, Yeh HJ, Haneef Z, Stern JM, Vannucci M. Bayesian vector autoregressive model for multi-subject effective connectivity inference using multi-modal neuroimaging data. Hum Brain Mapp 2017:38:1311-32.  Back to cited text no. 19
    
20.
Aguirre GK, Zarahn E, D'Esposito M. The variability of human, BOLD hemodynamic responses. Neuroimage 1998;8:360-9.  Back to cited text no. 20
    
21.
Handwerker DA, Ollinger JM, D'Esposito M. Variation of BOLD hemodynamic responses across subjects and brain regions and their effects on statistical analyses. Neuroimage 2004;21:1639-51.  Back to cited text no. 21
    
22.
Menz MM, Neumann J, Muller K, Zysset S. Variability of the BOLD response over time: An examination of within-session differences. Neuroimage 2006;32:1185-94.  Back to cited text no. 22
    
23.
Benar CG, Gross DW, Wang Y, Petre V, Pike B, Dubeau F, Gotman J. The BOLD response to interictal epileptiform discharges. Neuroimage 2002;17:1182-92.  Back to cited text no. 23
    
24.
Luders HO, Textbook of Epilepsy Surgery. 1st ed., London: CRC Press. 2008. p. 513-6.   Back to cited text no. 24
    
25.
Carreno M, Lüders, H.O., General principles of presurgical evaluations, in Epilepsy Surgery, H.O. Lüders, Comair, Y.G., Editor. 2001, Lippincott Williams & Wilkins: Philadelphia. p. 185-199.  Back to cited text no. 25
    
26.
Liu JV, Kobylarz EJ, Darcey TM, Lu Z, Wu YC, Meng M, Jobst BC. Improved mapping of interictal epileptiform discharges with EEG-fMRI and voxel-wise functional connectivity analysis. Epilepsia 2014;55:1380-8.  Back to cited text no. 26
    
27.
James GA, Tripathi SP, Ojemann JG, Gross RE, Drane DL. Diminished default mode network recruitment of the hippocampus and parahippocampus in temporal lobe epilepsy. J Neurosurg 2013;119:288-300.  Back to cited text no. 27
    
28.
Chiang S, Stern JM, Engel J, Jr., Levin HS, Haneef Z. Differences in graph theory functional connectivity in left and right temporal lobe epilepsy. Epilepsy Res 2014;108:1770-81.  Back to cited text no. 28
    
29.
Ridley BG, Rousseau C, Wirsich J, Le Troter A, Soulier E, Confort-Gouny S, et al. Nodal approach reveals differential impact of lateralized focal epilepsies on hub reorganization. Neuroimage 2015;118:39-48.  Back to cited text no. 29
    
30.
Haneef Z, Chiang S. Clinical correlates of graph theory findings in temporal lobe epilepsy. Seizure 2014;23:809-18.  Back to cited text no. 30
    
31.
Chiang S, Levin HS, Haneef Z. Computer-automated focus lateralization of temporal lobe epilepsy using fMRI. J Magn Reson Imaging 2015;41:1689-94.  Back to cited text no. 31
    
32.
Zhang Z, Lu G, Zhong Y, Tan Q, Chen H, Liao W, et al. fMRI study of mesial temporal lobe epilepsy using amplitude of low-frequency fluctuation analysis. Hum Brain Mapp 2010;31:1851-61.  Back to cited text no. 32
    
33.
Bettus G, Bartolomei F, Confort-Gouny S, Guedj E, Chauvel P, Cozzone PJ, et al. Role of resting state functional connectivity MRI in presurgical investigation of mesial temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 2010;81:1147-54.  Back to cited text no. 33
    
34.
Luo C, An D, Yao D, Gotman J. Patient-specific connectivity pattern of epileptic network in frontal lobe epilepsy. Neuroimage Clin 2014;4:668-75.  Back to cited text no. 34
    
35.
Stufflebeam SM, Liu H, Sepulcre J, Tanaka N, Buckner RL, Madsen JR. Localization of focal epileptic discharges using functional connectivity magnetic resonance imaging. J Neurosurg 2011;114:1693-7.  Back to cited text no. 35
    
36.
Nedic S, Stufflebeam SM, Rondinoni C, Velasco TR, dos Santos AC, Leite JP, et al. Using network dynamic fMRI for detection of epileptogenic foci. BMC Neurol 2015;15:262.  Back to cited text no. 36
    
37.
Weaver KE, Chaovalitwongse WA, Novotny EJ, Poliakov A, Grabowski TG, Ojemann JG. Local functional connectivity as a pre-surgical tool for seizure focus identification in non-lesion, focal epilepsy. Front Neurol 2013;4:43.  Back to cited text no. 37
    
38.
Schevon CA, Cappell J, Emerson R, Isler J, Grieve P, Goodman R, et al. Cortical abnormalities in epilepsy revealed by local EEG synchrony. Neuroimage 2007;35:140-8.  Back to cited text no. 38
    
39.
Bettus G, Wendling F, Guye M, Valton L, Regis J, Chauvel P, et al. Enhanced EEG functional connectivity in mesial temporal lobe epilepsy. Epilepsy Res 2008;81:58-68.  Back to cited text no. 39
    
40.
Honey CJ, Sporns O, Cammoun L, Gigandet X, Thiran JP, Meuli R, et al. Predicting human resting-state functional connectivity from structural connectivity. Proc Natl Acad Sci U S A 2009;106:2035-40.  Back to cited text no. 40
    
41.
Ma S, Calhoun VD, Phlypo R, Adali T. Dynamic changes of spatial functional network connectivity in healthy individuals and schizophrenia patients using independent vector analysis. Neuroimage 2014;90:196-206.  Back to cited text no. 41
    
42.
Chiang S, Cassese A, Guindani M, Vannucci M, Yeh HJ, Haneef Z, Stern JM. Time-dependence of graph theory metrics in functional connectivity analysis. Neuroimage 2016;125:601-15.  Back to cited text no. 42
    
43.
Negishi M, Martuzzi R, Novotny EJ, Spencer DD, Constable RT. Functional MRI connectivity as a predictor of the surgical outcome of epilepsy. Epilepsia 2011;52:1733-40.  Back to cited text no. 43
    
44.
Lee HW, Arora J, Papademetris X, Tokoglu F, Negishi M, Scheinost D, Farooque P, Blumenfeld H, Spencer DD, Constable RT. Altered functional connectivity in seizure onset zones revealed by fMRI intrinsic connectivity. Neurology 2014;83:2269-77.  Back to cited text no. 44
    
45.
Haneef Z, Chiang S, Yeh HJ, Engel J, Jr., Stern JM. Functional connectivity homogeneity correlates with duration of temporal lobe epilepsy. Epilepsy Behav 2015;46:227-33.  Back to cited text no. 45
    
46.
Morgan VL, Abou-Khalil B, Rogers BP. Evolution of functional connectivity of brain networks and their dynamic interaction in temporal lobe epilepsy. Brain Connect 2015;5:35-44.  Back to cited text no. 46
    
47.
Xiong J, Parsons LM, Gao JH, Fox PT. Interregional connectivity to primary motor cortex revealed using MRI resting state images. Hum Brain Mapp 1999;8:151-6.  Back to cited text no. 47
    
48.
Zhang D, Johnston JM, Fox MD, Leuthardt EC, Grubb RL, Chicoine MR, et al. Preoperative sensorimotor mapping in brain tumor patients using spontaneous fluctuations in neuronal activity imaged with functional magnetic resonance imaging: Initial experience. Neurosurgery 2009;65(6 Suppl):226-36.  Back to cited text no. 48
    
49.
Liu H, Buckner RL, Talukdar T, Tanaka N, Madsen JR, Stufflebeam SM. Task-free presurgical mapping using functional magnetic resonance imaging intrinsic activity. J Neurosurg 2009;111:746-54.  Back to cited text no. 49
    
50.
McKeown MJ, Makeig S, Brown GG, Jung TP, Kindermann SS, Bell AJ, et al. Analysis of fMRI data by blind separation into independent spatial components. Hum Brain Mapp 1998;6:160-88.  Back to cited text no. 50
    
51.
Allen EA, Erhardt EB, Damaraju E, Gruner W, Segall JM, Silva RF, et al. A baseline for the multivariate comparison of resting-state networks. Front Syst Neurosci 2011;5:2.  Back to cited text no. 51
    
52.
Beckmann CF, DeLuca M, Devlin JT, Smith SM. Investigations into resting-state connectivity using independent component analysis. Philos Trans R Soc Lond B Biol Sci 2005;360:1001-13.  Back to cited text no. 52
    
53.
Damoiseaux JS, Rombouts SA, Barkhof F, Scheltens P, Stam CJ, Smith SM, Beckmann CF. Consistent resting-state networks across healthy subjects. Proc Natl Acad Sci U S A 2006;103:13848-53.  Back to cited text no. 53
    
54.
De Luca M, Beckmann CF, De Stefano N, Matthews PM, Smith SM. fMRI resting state networks define distinct modes of long-distance interactions in the human brain. Neuroimage 2006;29:1359-67.  Back to cited text no. 54
    
55.
Kiviniemi V, Kantola JH, Jauhiainen J, Hyvarinen A, Tervonen O. Independent component analysis of nondeterministic fMRI signal sources. Neuroimage 2003;19(2 Pt 1):253-60.  Back to cited text no. 55
    
56.
van de Ven VG, Formisano E, Prvulovic D, Roeder CH, Linden DE. Functional connectivity as revealed by spatial independent component analysis of fMRI measurements during rest. Hum Brain Mapp 2004;22:165-78.  Back to cited text no. 56
    
57.
Kokkonen SM, Nikkinen J, Remes J, Kantola J, Starck T, Haapea M, et al. Preoperative localization of the sensorimotor area using independent component analysis of resting-state fMRI. Magn Reson Imaging 2009;27:733-40.  Back to cited text no. 57
    
58.
Mitchell TJ, Hacker CD, Breshears JD, Szrama NP, Sharma M, Bundy DT, et al. A novel data-driven approach to preoperative mapping of functional cortex using resting-state functional magnetic resonance imaging. Neurosurgery 2013;73:969-83.  Back to cited text no. 58
    
59.
Ma L, Wang B, Chen X, Xiong J. Detecting functional connectivity in the resting brain: A comparison between ICA and CCA. Magn Reson Imaging 2007;25:47-56.  Back to cited text no. 59
    
60.
Greicius MD, Srivastava G, Reiss AL, Menon V. Default-mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI. Proc Natl Acad Sci U S A 2004;101:4637-42.  Back to cited text no. 60
    
61.
Calhoun VD, Adali T, McGinty VB, Pekar JJ, Watson TD, Pearlson GD. fMRI activation in a visual-perception task: Network of areas detected using the general linear model and independent components analysis. Neuroimage 2001;14:1080-8.  Back to cited text no. 61
    
62.
van de Ven V, Bledowski C, Prvulovic D, Goebel R, Formisano E, Di Salle F, Linden DE, Esposito F. Visual target modulation of functional connectivity networks revealed by self-organizing group ICA. Hum Brain Mapp 2008;29:1450-61.  Back to cited text no. 62
    
63.
Sanai N, Mirzadeh Z, Berger MS. Functional outcome after language mapping for glioma resection. N Engl J Med 2008;358:18-27.  Back to cited text no. 63
    
64.
Tie Y, Rigolo L, Norton IH, Huang RY, Wu W, Orringer D, et al. Defining language networks from resting-state fMRI for surgical planning-a feasibility study. Hum Brain Mapp 2014;35:1018-30.  Back to cited text no. 64
    
65.
Van Essen DC, Smith SM, Barch DM, Behrens TE, Yacoub E, Ugurbil K. The WU-Minn Human Connectome Project: An overview. Neuroimage 2013;80:62-79.  Back to cited text no. 65
    
66.
Holmes AJ, Hollinshead MO, O'Keefe TM, Petrov VI, Fariello GR, Wald LL, Fischl B, Rosen BR, Mair RW, Roffman JL, Smoller JW, Buckner RL. Brain Genomics Superstruct Project initial data release with structural, functional, and behavioral measures. Sci Data 2015;2:150031.  Back to cited text no. 66
    
67.
Branco P, Seixas D, Deprez S, Kovacs S, Peeters R, Castro SL, et al. Resting-state functional magnetic resonance imaging for language preoperative planning. Front Hum Neurosci 2016;10:11.  Back to cited text no. 67
    


    Figures

  [Figure 1], [Figure 2]
 
 
    Tables

  [Table 1], [Table 2]



 

Top
Print this article  Email this article
   
Online since 20th March '04
Published by Wolters Kluwer - Medknow