Recursive grid partitioning on a cortical surface model: an optimized technique for the localization of implanted subdural electrodes

Clinical article

Restricted access

Object

Precise localization of subdural electrodes (SDEs) is essential for the interpretation of data from intracranial electrocorticography recordings. Blood and fluid accumulation underneath the craniotomy flap leads to a nonlinear deformation of the brain surface and of the SDE array on postoperative CT scans and adversely impacts the accurate localization of electrodes located underneath the craniotomy. Older methods that localize electrodes based on their identification on a postimplantation CT scan with coregistration to a preimplantation MR image can result in significant problems with accuracy of the electrode localization. The authors report 3 novel methods that rely on the creation of a set of 3D mesh models to depict the pial surface and a smoothed pial envelope. Two of these new methods are designed to localize electrodes, and they are compared with 6 methods currently in use to determine their relative accuracy and reliability.

Methods

The first method involves manually localizing each electrode using digital photographs obtained at surgery. This is highly accurate, but requires time intensive, operator-dependent input. The second uses 4 electrodes localized manually in conjunction with an automated, recursive partitioning technique to localize the entire electrode array. The authors evaluated the accuracy of previously published methods by applying the methods to their data and comparing them against the photograph-based localization. Finally, the authors further enhanced the usability of these methods by using automatic parcellation techniques to assign anatomical labels to individual electrodes as well as by generating an inflated cortical surface model while still preserving electrode locations relative to the cortical anatomy.

Results

The recursive grid partitioning had the least error compared with older methods (672 electrodes, 6.4-mm maximum electrode error, 2.0-mm mean error, p < 10−18). The maximum errors derived using prior methods of localization ranged from 8.2 to 11.7 mm for an individual electrode, with mean errors ranging between 2.9 and 4.1 mm depending on the method used. The authors also noted a larger error in all methods that used CT scans alone to localize electrodes compared with those that used both postoperative CT and postoperative MRI. The large mean errors reported with these methods are liable to affect intermodal data comparisons (for example, with functional mapping techniques) and may impact surgical decision making.

Conclusions

The authors have presented several aspects of using new techniques to visualize electrodes implanted for localizing epilepsy. The ability to use automated labeling schemas to denote which gyrus a particular electrode overlies is potentially of great utility in planning resections and in corroborating the results of extraoperative stimulation mapping. Dilation of the pial mesh model provides, for the first time, a sense of the cortical surface not sampled by the electrode, and the potential roles this “electrophysiologically hidden” cortex may play in both eloquent function and seizure onset.

Abbreviations used in this paper:AC-PC = anterior commissure–posterior commissure; CSM = cortical stimulation mapping; fMRI = functional MRI; MEG = magnetoencephalography; ROI = region of interest; SDE = subdural electrode.

Article Information

Address correspondence to: Nitin Tandon, M.D., University of Texas Medical School at Houston, 6431 Fannin Street, Suite G.550, Houston, Texas 77030. email: nitin.tandon@uth.tmc.edu.

Please include this information when citing this paper: published online March 15, 2013; DOI: 10.3171/2013.2.JNS121450.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Intraoperative photograph–3D pial model colocalization (Method 1). A: Comparison of gyral and sulcal landmarks that provide the connection between the 2D photograph and the 3D surface volume. The blue lines in panel A match those in panel B. B: Localization of target electrodes in one of the patients. The localizations of the electrodes in the explantation photograph are compared with those in the intraoperative photograph. C: Final locations are easily visualized on the high-resolution 3D cortical surface. The yellow line marks the boundary of the craniotomy.

  • View in gallery

    Recursive grid partitioning technique (Method 2). The process of electrode localization is based on 4 surface coordinates chosen from intraoperative photographs. Upper: Coordinates of the 4 corners of the grid are chosen based on the intraoperative photograph compared with a 3D pial mesh model generated using a high-resolution T1-weighted image. Custom MATLAB Script is used to create lines on the 3D cortical surface connecting these locations. This grid is then portioned repeatedly until the program arrives at the appropriate numbers of rows and columns. Electrodes are then localized centered at grid intersections to complete the process. Lower: The extrapolation technique is used to test the accuracy of this technique. On a grid entirely visible through the craniotomy, a subset of electrodes (red) are chosen and localized on the pial surface using anatomical landmarks. Intervening electrodes (green) are localized and used to extrapolate the “nonvisible” electrodes (blue). Given that these electrodes are in fact visible on the intraoperative photograph (inset), their estimated locations could be compared with their actual ones. The average error of electrodes in green was 1.75 mm and of the electrodes in blue was 1.99 mm.

  • View in gallery

    Illustration of conventional methods of dilation and errors. A–C: Post-SDE implantation CT scan showing displacement of electrodes due to fluid accumulation after skull flap replacement (A). High-resolution preimplantation T1-weighted MRI of the same axial slice with the pial surface in blue and the pial-enveloped surface in red (B). The CT is shown as an overlay to illustrate the severity of the shift produced by the electrodes and the potential for mislocalization (C). D–F: Artifact method. The CT scan is depicted on the pial surface as a thresholded overlay. Electrode locations are easily chosen as the intersection of high-intensity artifact (yellow and orange) with the pial surface (D). These artifacts are used to localize the electrodes (blue; E), which can then be compared with their actual location (white; F). G–I: Dilation method using the midpoint of the AC-PC line. The electrodes are chosen from the center of mass of their CT artifact thoroughly “buried” inside the pial surface (G). The same electrodes are shown dilated out to the enveloped pial surface (H). Finally, these electrodes (red) are compared with their actual locations based on the intraoperative photograph localization method (white; I).

  • View in gallery

    Parcellation and lissencephalic surface generation. A representation of both the parcellation method and inflation method is shown here. The upper panel shows a patient's pial surface with both a frontal and a temporal grid represented. The automated parcellation process outlines edges between particular sulci and gyri, which are represented using different colors. The parcellation scheme is shown in the lower panel on the dilated lissencephalic pial surface with the electrodes still placed in their anatomical positions using ROIs placed around each electrode in the natural space that are then transformed along with the surface dilation. This shows how apparently adjoining electrodes may have large amounts of gray matter intervening between them, which could be factored into the interpretation of data from them regarding seizure onset zones as well as functional mapping data. Colors that are not shown in the legend are either irrelevant (no electrodes localized to them) or designate sulci. The numbers represent the electrode numbers on each SDE array.

  • View in gallery

    Surface mapping concordance. Electrodes that are found to overlap eloquent cortical sites (positive cortical stimulation mapping electrodes) along with all the other electrodes implanted in a patient are depicted in 4 patients (the same patients shown in Fig. 1). Electrodes represented by black dots produced deficits when stimulated during language tasks. Electrodes represented by white dots either produced no deficits or were not tested due to lack of clinical relevance. It can be seen that these electrodes overlie the anterior and posterior language sites, and prior knowledge about the relationship of individual electrodes to cortical anatomy (for example, pars opercularis or posterior superior temporal gyrus) may allow for more directed efforts at functional localization and surgical planning. As in Fig. 4, labels for each color are given to the right. Colors that are not shown in the legend are either irrelevant (no electrodes localized to them) or designate sulci.

  • View in gallery

    Comparative assessment of errors of various localization techniques. Boxplots showing the distribution of the errors (along the y axis) of each method (along the x axis) compared with the intraoperative photograph localization method. For each boxplot, the middle line with the notched box represents the median value, while the edges of the box represent the 25th and 75th percentiles. The whiskers extend to the extremes that are not considered outliers. Outliers are depicted as plus signs. A single asterisk indicates a method that was better than at least 1 of the other methods. Double asterisks indicate significance over all other methods except for recursive grid partitioning method. Triple asterisks indicate significance of a given method over all other methods. A cutoff of p < 0.001 was used as the significance threshold.

  • View in gallery

    Comparison of errors of 4 methods against the intraoperative photograph localization technique. Electrodes localized via Methods 1–5 for a temporal grid on a single patient. White dots represent electrodes in Method 1 (intraoperative photograph localization method). Turquoise dots represent electrodes localized using Method 2 (recursive grid partitioning). Green dots represent electrodes based on Method 3 (artifact method). Red dots represent electrodes localized by Method 4 (dilation method). Blue dots represent electrodes localized by Method 5 (surface normal method).

References

  • 1

    Cascino GDBuchhalter JRMullan BPSo EL: Ictal SPECT in nonlesional extratemporal epilepsy. Epilepsia 45:Suppl 432342004

  • 2

    Chang EFRieger JWJohnson KBerger MSBarbaro NMKnight RT: Categorical speech representation in human superior temporal gyrus. Nat Neurosci 13:142814322010

    • Search Google Scholar
    • Export Citation
  • 3

    Conner CREllmore TMDiSano MAPieters TAPotter AWTandon N: Anatomic and electro-physiologic connectivity of the language system: a combined DTI-CCEP study. Comput Biol Med 41:110011092011

    • Search Google Scholar
    • Export Citation
  • 4

    Conner CREllmore TMPieters TADiSano MATandon N: Variability of the relationship between electrophysiology and BOLD-fMRI across cortical regions in humans. J Neurosci 31:12855128652011

    • Search Google Scholar
    • Export Citation
  • 5

    Cox RW: AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res 29:1621731996

    • Search Google Scholar
    • Export Citation
  • 6

    Crone NEMiglioretti DLGordon BLesser RP: Functional mapping of human sensorimotor cortex with electrocorticographic spectral analysis. II. Event-related synchronization in the gamma band. Brain 121:230123151998

    • Search Google Scholar
    • Export Citation
  • 7

    Dalal SSEdwards EKirsch HEBarbaro NMKnight RTNagarajan SS: Localization of neurosurgically implanted electrodes via photograph-MRI-radiograph coregistration. J Neurosci Methods 174:1061152008

    • Search Google Scholar
    • Export Citation
  • 8

    Dale AMFischl BSereno MI: Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9:1791941999

  • 9

    Diehl BPiao ZTkach JBusch RMLaPresto ENajm I: Cortical stimulation for language mapping in focal epilepsy: correlations with tractography of the arcuate fasciculus. Epilepsia 51:6396462010

    • Search Google Scholar
    • Export Citation
  • 10

    Dykstra ARChan AMQuinn BTZepeda RKeller CJCormier J: Individualized localization and cortical surface-based registration of intracranial electrodes. Neuroimage 59:356335702012

    • Search Google Scholar
    • Export Citation
  • 11

    Edwards ENagarajan SSDalal SSCanolty RTKirsch HEBarbaro NM: Spatiotemporal imaging of cortical activation during verb generation and picture naming. Neuroimage 50:2913012010

    • Search Google Scholar
    • Export Citation
  • 12

    Elias WJFu KMFrysinger RC: Cortical and subcortical brain shift during stereotactic procedures. J Neurosurg 107:9839882007

  • 13

    Engel AKMoll CKFried IOjemann GA: Invasive recordings from the human brain: clinical insights and beyond. Nat Rev Neurosci 6:35472005

    • Search Google Scholar
    • Export Citation
  • 14

    Fischl BSereno MIDale AM: Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. Neuroimage 9:1952071999

    • Search Google Scholar
    • Export Citation
  • 15

    Fischl Bvan der Kouwe ADestrieux CHalgren ESégonne FSalat DH: Automatically parcellating the human cerebral cortex. Cereb Cortex 14:11222004

    • Search Google Scholar
    • Export Citation
  • 16

    Flinker AChang EFBarbaro NMBerger MSKnight RT: Sub-centimeter language organization in the human temporal lobe. Brain Lang 117:1031092011

    • Search Google Scholar
    • Export Citation
  • 17

    Hastreiter PRezk-Salama CSoza GBauer MGreiner GFahlbusch R: Strategies for brain shift evaluation. Med Image Anal 8:4474642004

    • Search Google Scholar
    • Export Citation
  • 18

    Hermes DMiller KJNoordmans HJVansteensel MJRamsey NF: Automated electrocorticographic electrode localization on individually rendered brain surfaces. J Neurosci Methods 185:2932982010

    • Search Google Scholar
    • Export Citation
  • 19

    Hill DLSmith ADSimmons AMaurer CR JrCox TCElwes R: Sources of error in comparing functional magnetic resonance imaging and invasive electrophysiological recordings. J Neurosurg 93:2142232000

    • Search Google Scholar
    • Export Citation
  • 20

    Hunter JDHanan DMSinger BFShaikh SBrubaker KAHecox KE: Locating chronically implanted subdural electrodes using surface reconstruction. Clin Neurophysiol 116:198419872005

    • Search Google Scholar
    • Export Citation
  • 21

    Ishibashi HSimos PGCastillo EMMaggio WWWheless JWKim HL: Detection and significance of focal, interictal, slow-wave activity visualized by magnetoencephalography for localization of a primary epileptogenic region. J Neurosurg 96:7247302002

    • Search Google Scholar
    • Export Citation
  • 22

    Kovalev DSpreer JHonegger JZentner JSchulze-Bonhage AHuppertz HJ: Rapid and fully automated visualization of subdural electrodes in the presurgical evaluation of epilepsy patients. AJNR Am J Neuroradiol 26:107810832005

    • Search Google Scholar
    • Export Citation
  • 23

    Lachaux JPChavez MLutz A: A simple measure of correlation across time, frequency and space between continuous brain signals. J Neurosci Methods 123:1751882003

    • Search Google Scholar
    • Export Citation
  • 24

    Lachaux JPFonlupt PKahane PMinotti LHoffmann DBertrand O: Relationship between task-related gamma oscillations and BOLD signal: new insights from combined fMRI and intracranial EEG. Hum Brain Mapp 28:136813752007

    • Search Google Scholar
    • Export Citation
  • 25

    Lachaux JPGeorge NTallon-Baudry CMartinerie JHugueville LMinotti L: The many faces of the gamma band response to complex visual stimuli. Neuroimage 25:4915012005

    • Search Google Scholar
    • Export Citation
  • 26

    Lachaux JPJerbi KBertrand OMinotti LHoffmann DSchoendorff B: A blueprint for real-time functional mapping via human intracranial recordings. PLoS ONE 2:e10942007

    • Search Google Scholar
    • Export Citation
  • 27

    Lachaux JPRudrauf DKahane P: Intracranial EEG and human brain mapping. J Physiol Paris 97:6136282003

  • 28

    Lesser RPCrone NEWebber WR: Subdural electrodes. Clin Neurophysiol 121:137613922010

  • 29

    Matsumoto RNair DRLaPresto ENajm IBingaman WShibasaki H: Functional connectivity in the human language system: a cortico-cortical evoked potential study. Brain 127:231623302004

    • Search Google Scholar
    • Export Citation
  • 30

    Miller KJLeuthardt ECSchalk GRao RPAnderson NRMoran DW: Spectral changes in cortical surface potentials during motor movement. J Neurosci 27:242424322007

    • Search Google Scholar
    • Export Citation
  • 31

    Miller KJMakeig SHebb AORao RPdenNijs MOjemann JG: Cortical electrode localization from X-rays and simple mapping for electrocorticographic research: The “Location on Cortex” (LOC) package for MATLAB. J Neurosci Methods 162:3033082007

    • Search Google Scholar
    • Export Citation
  • 32

    Morris KO'Brien TJCook MJMurphy MBowden SC: A computer-generated stereotactic “Virtual Subdural Grid” to guide resective epilepsy surgery. AJNR Am J Neuroradiol 25:77832004

    • Search Google Scholar
    • Export Citation
  • 33

    O'Shea JPWells WM IIIGolby AJ: Using surface normals to localize subdural intracranial electrodes placed during neurosurgery. Proc IEEE Int Symp Biomed Imaging 3313342006

    • Search Google Scholar
    • Export Citation
  • 34

    Rutka JTOtsubo HKitano SSakamoto HShirasawa AOchi A: Utility of digital camera-derived intraoperative images in the planning of epilepsy surgery for children. Neurosurgery 45:118611911999

    • Search Google Scholar
    • Export Citation
  • 35

    Saad ZSRopella KMDeYoe EABandettini PA: The spatial extent of the BOLD response. Neuroimage 19:1321442003

  • 36

    Schulze-Bonhage AHHuppertz HJComeau RMHonegger JBSpreer JMZentner JK: Visualization of subdural strip and grid electrodes using curvilinear reformatting of 3D MR imaging data sets. AJNR Am J Neuroradiol 23:4004032002

    • Search Google Scholar
    • Export Citation
  • 37

    Sebastiano FDi Gennaro GEsposito VPicardi AMorace RSparano A: A rapid and reliable procedure to localize subdural electrodes in presurgical evaluation of patients with drug-resistant focal epilepsy. Clin Neurophysiol 117:3413472006

    • Search Google Scholar
    • Export Citation
  • 38

    Simos PGPapanicolaou ACBreier JIWheless JWConstantinou JEGormley WB: Localization of language-specific cortex by using magnetic source imaging and electrical stimulation mapping. J Neurosurg 91:7877961999

    • Search Google Scholar
    • Export Citation
  • 39

    Škrinjar ODuncan J: Preserving intrinsic surface distances—application to electrode grid manipulation. Proc IEEE Workshop Math Methods Biomed Image Analysis 54602000

    • Search Google Scholar
    • Export Citation
  • 40

    Škrinjar ONabavi ADuncan J: Model-driven brain shift compensation. Med Image Anal 6:3613732002

  • 41

    Studholme CNovotny EZubal IGDuncan JS: Estimating tissue deformation between functional images induced by intracranial electrode implantation using anatomical MRI. Neuroimage 13:5615762001

    • Search Google Scholar
    • Export Citation
  • 42

    Swann NPoizner HHouser MGould SGreenhouse ICai W: Deep brain stimulation of the subthalamic nucleus alters the cortical profile of response inhibition in the beta frequency band: a scalp EEG study in Parkinson's disease. J Neurosci 31:572157292011

    • Search Google Scholar
    • Export Citation
  • 43

    Swann NTandon NCanolty REllmore TMMcEvoy LKDreyer S: Intracranial EEG reveals a time- and frequency-specific role for the right inferior frontal gyrus and primary motor cortex in stopping initiated responses. J Neurosci 29:12675126852009

    • Search Google Scholar
    • Export Citation
  • 44

    Swann NCCai WConner CRPieters TAClaffey MPGeorge JS: Roles for the pre-supplementary motor area and the right inferior frontal gyrus in stopping action: electrophysiological responses and functional and structural connectivity. Neuroimage 59:286028702012

    • Search Google Scholar
    • Export Citation
  • 45

    Tandon NCortical mapping by electrical stimulation of subdural electrodes: language areas. Lüders HO: Textbook of Epilepsy Surgery LondonInforma Healthcare2008. 10011015

    • Search Google Scholar
    • Export Citation
  • 46

    Tandon NMapping of human language. Yoshor DMizrahi E: Clinical Brain Mapping New YorkMcGraw Hill2012. 203218

  • 47

    Tao JXHawes-Ebersole SBaldwin MShah SErickson RKEbersole JS: The accuracy and reliability of 3D CT/MRI coregistration in planning epilepsy surgery. Clin Neurophysiol 120:7487532009

    • Search Google Scholar
    • Export Citation
  • 48

    Wang YAgarwal RNguyen DDomocos VGotman J: Intracranial electrode visualization in invasive pre-surgical evaluation for epilepsy. Conf Proc IEEE Eng Med Biol Soc 1:9529552005

    • Search Google Scholar
    • Export Citation
  • 49

    Wang ZIJin KKakisaka YMosher JCBingaman WEKotagal P: Imag(in)ing seizure propagation: MEG-guided interpretation of epileptic activity from a deep source. Hum Brain Mapp 33:279728012012

    • Search Google Scholar
    • Export Citation
  • 50

    Wellmer Jvon Oertzen JSchaller CUrbach HKönig RWidman G: Digital photography and 3D MRI-based multimodal imaging for individualized planning of resective neocortical epilepsy surgery. Epilepsia 43:154315502002

    • Search Google Scholar
    • Export Citation
  • 51

    Winkler PAVollmar CKrishnan KGPfluger TBrückmann HNoachtar S: Usefulness of 3-D reconstructed images of the human cerebral cortex for localization of subdural electrodes in epilepsy surgery. Epilepsy Res 41:1691782000

    • Search Google Scholar
    • Export Citation

TrendMD

Cited By

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 225 225 24
Full Text Views 295 258 3
PDF Downloads 124 108 1
EPUB Downloads 0 0 0

PubMed

Google Scholar