UBC Faculty Research and Publications

Shape (but not volume) changes in the thalami in Parkinson disease McKeown, Martin J; Uthama, Ashish; Abugharbieh, Rafeef; Palmer, Samantha; Lewis, Mechelle; Huang, Xuemei Apr 16, 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12883_2007_Article_170.pdf [ 1.02MB ]
Metadata
JSON: 52383-1.0167789.json
JSON-LD: 52383-1.0167789-ld.json
RDF/XML (Pretty): 52383-1.0167789-rdf.xml
RDF/JSON: 52383-1.0167789-rdf.json
Turtle: 52383-1.0167789-turtle.txt
N-Triples: 52383-1.0167789-rdf-ntriples.txt
Original Record: 52383-1.0167789-source.json
Full Text
52383-1.0167789-fulltext.txt
Citation
52383-1.0167789.ris

Full Text

ralssBioMed CentBMC NeurologyOpen AcceResearch articleShape (but not volume) changes in the thalami in Parkinson diseaseMartin J McKeown*†1,2,3,4, Ashish Uthama†4, Rafeef Abugharbieh†2,4, Samantha Palmer†1, Mechelle Lewis†5 and Xuemei Huang†5Address: 1Pacific Parkinson's Research Center, University of British Columbia, Vancouver, Canada, 2Brain Research Centre, University of British Columbia, Vancouver, Canada, 3Department of Medicine (Neurology), University of British Columbia, Vancouver, Canada, 4Biomedical Signal and Image Computing Lab, Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada and 5Department of Neurology, University of North Carolina, Chapel Hill, NC, USAEmail: Martin J McKeown* - mmckeown@interchange.ubc.ca; Ashish Uthama - ashishu@ece.ubc.ca; Rafeef Abugharbieh - rafeef@ece.ubc.ca; Samantha Palmer - sjpalmer@interchange.ubc.ca; Mechelle Lewis - lewism@neurology.unc.edu; Xuemei Huang - xuemei@med.unc.edu* Corresponding author    †Equal contributorsAbstractBackground: Recent pathological studies have suggested that thalamic degeneration mayrepresent a site of non-dopaminergic degeneration in Parkinson's Disease (PD). Our objective wasto determine if changes in the thalami could be non-invasively detected in structural MRI imagesobtained from subjects with Parkinson disease (PD), compared to age-matched controls.Results: No significant differences in volume were detected in the thalami between eighteennormal subjects and eighteen PD subjects groups. However significant (p < 0.03) shape differenceswere detected between the Left vs. Right thalami in PD, between the left thalami in PD andcontrols, and between the right thalami in PD and controls using a recently-developed, sphericalharmonic-based representation.Conclusion: Systematic changes in thalamic shape can be non-invasively assessed in PD in vivo.Shape changes, in addition to volume changes, may represent a new avenue to assess the progressof neurodegenerative processes. Although not directly discernable at the resolution of standardMRI, previous pathological studies would suggest that the shape changes detected in this studyrepresent degeneration in the centre median-parafascicular (CM-Pf) complex, an area known torepresent selective non-dopaminergic degeneration in PD.BackgroundThe thalamic changes seen in Parkinson Disease (PD)may represent selective non-dopaminergic degeneration[1], as there is selective neuronal loss in the centremedian-parafascicular (CM-Pf) complex in Parkinson'sdisease [2], yet relative preservation of neurons in the lim-bic (mediodorsal and anterior principal) thalamic nuclei.As expected, they found α-synuclein-positive Lewy bodiesin these nuclei in the thalamus, but they also found a sig-nificant reduction (40–55%) in the total neuronalnumber in the caudal intralaminar (CM-Pf) nuclei,regions that receive glutaminergic innervation [3]. Thiscontrasted with the 70% loss of pigmented nigral neu-rons. A factor analysis has demonstrated that the size ofPublished: 16 April 2008BMC Neurology 2008, 8:8 doi:10.1186/1471-2377-8-8Received: 4 October 2007Accepted: 16 April 2008This article is available from: http://www.biomedcentral.com/1471-2377/8/8© 2008 McKeown et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 8(page number not for citation purposes)Henderson et al. examined the thalamic intranuclearnuclei in 10 normal controls and 9 patients with PD [3].neurons in the motor cortex is negatively correlated withthe size and number of neurons in its thalamic relay, theBMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8VLp. There is also a positive correlation between thenumber of ventral anterior (VA) neurons and the pre-sup-plementary motor area (SMA) [4].Bacci et al. suggested that CM-Pf degeneration may par-tially counteract the consequences of dopamine neuronalloss, as thalamic and dopamine inputs have antagonisticinfluence on neurotransmitter-related gene expression[1]. Moreover, the CM-Pf degeneration may be a directconsequence of nigrostriatal denervation, as depleting thestriatum of dopamine results in the remaining Pf neuronsbeing particularly hyperactive [5].The normal role of the CM-Pf complex is incompletelyunderstood, although it is clearly related to basal gangliafunction [6]. The Pf nuclei receive input from the spinalcord and project to the striatum [7]. These projectionsmay carry specific temporally-patterned inputs to striataltargets [8]. While the CM-Pf complex has traditionallybeen considered part of the reticulo-thalamo-cortical acti-vating system, a recent proposal suggests that the CM-Pfcomplex participates in sensory driven attentional proc-esses, particularly unexpected events [9].The advent of modern imaging techniques has allowedthe noninvasive in vivo assessment of brain structures,such as the thalamus, in disease states. Thalamic morpho-logical changes have been detected chronically after corti-cal injury, such as middle cerebral artery (MCA) infarctionafter 1 yr [10,11] and tumor resection after ~2 yrs [12].The study by Hulshoff Pol [12] detected a 5% decrease inipsilateral thalamus and, interestingly, a 4.5% increase incontralateral thalamic volume after unilateral tumorresection, presumably on the basis of a compensatorymechanism.In PD and related disorders, some studies of structuraland functional imaging have detected thalamic changes.Thalamic grey matter changes have been detected contral-ateral to unilateral Parkinsonian resting tremor [13]. InPD with dementia, the thalamus, in addition to the hip-pocampus and anterior cingulate, represent the regionsmost affected [14]. Functionally, there is a connectionbetween a major component in F-DOPA uptake in thestriatum and a component from fluoro-deoxy – glucose(FDG), which had positive loadings in the thalamus andthe cerebellum [15].Most morphological studies based on imaging involve anumber of steps manipulating the brain images. A typicalapproach would be to warp ("spatially normalize") thebrain images to a common space [13]. Further smoothingof the data (e.g. using an isotropic 12 mm Gaussian ker-inferences about small, subcortical structures like the tha-lamus. In fact this "Voxel Based Morphometry" approachhas thus been a controversial approach (e.g., see discus-sions in [16] and [17]).Recent approaches try to reduce errors due to misregistra-tion by aligning the subjects at the region of interest (ROI)level, as opposed to the whole brain level [18-20]. How-ever, these approaches are designed to deal with a differ-ent problem, namely that of summarizing fMRI activationfrom several subjects. To quantify differences in morphol-ogy, it would be necessary to examine the different trans-formations required to warp each subject's ROIs to theexamplar ROI shape – a non-trivial task.An alternative approach to warping brains to the samespace is to segment brain structures individually onunmanipulated (i.e. unregistered and unwarped) brains[21]. Because no registration of the brain images is done,this requires summarizing the individual brain structuresin a way that they are invariant to positioning of the headin the scanner. For example, simply estimating the vol-ume of an ROI such as the thalamus has this property, asit is invariant to the individual coordinate frame used. Anumber of such invariants (e.g. spatial variance) havebeen proposed to summarize the shape of brain structures[22], or even characterize the distribution (texture) of acti-vation maps in fMRI [23].We have recently proposed a method based on sphericalharmonics (SPHARM) which provided a unique represen-tation of brain structures, including regions with possibletopological disconnections, such as the lateral ventricles[24]. In brief, the method involves first finding a sphericalshell which encompasses the entire ROI. Subsequentsmaller concentric shells are then derived and the intersec-tion between the progressively smaller spherical shellsand the brain structure is computed (Figure 1). The resultsfrom the intersections are then combined into a uniquefeature vector containing approximately 100's or 1000's ofelements. This feature vector provides a unique represen-tation of the shape which is independent of the spatial ori-entation of the structure (see Methods).We examined the thalami from 18 PD subjects and 18age-matched controls. Using the above technique, wefound significant differences between the two groups inthe shape of the thalami, but not in the volume. This sug-gests that significant thalamic changes can be assessednoninvasively in PD, suitable for longitudinal studies.ResultsThere were no significant differences in volume betweenPage 2 of 8(page number not for citation purposes)nel) to minimize the effects of misregistration betweendifferent normalized brains may affect the ability to makesides in either controls or PD subjects, nor between con-trols and PD subjects in either the left or right thalamusBMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8(Table 1). In contrast, the SPHARM-based method foundsignificant shape differences between the left and rightthalamus in PD but not in controls. Significant shape dif-ferences between PD and controls were detected in boththe left and right thalami.For PD subjects, an ANOVA examining the influences onthe distance metric (Eqn 8) of acquisition site, side ofsymptoms or duration (F(1,20) = 1.9, p = 0.18; F(1,20) =1.59, p = 0.22; F(17,20) = 1.35, p = 0.26; respectively). Asall PD subjects were right handed, handedness could notbe tested in this group. Similarly, an ANOVA on the dis-tance metric for all subjects examining site of acquisition(F(1,72) = 0.14, p = 0.71) or handedness (F(1,72) = 0.16,p = 0.69) did not find factors of significance with a fixedeffects model.In order to better visualize and intuitively assess the shapedifferences in thalami between PD subjects and controls,we took thalami that had "typical" feature vectors (i.e. fea-ture vectors closest to the mean of each group) andassumed that they represented exemplar shapes. We thenspatially aligned these exemplar shapes (Figure 2). Thereappeared to be greater differences in the left thalamibetween controls and PD subjects. The largest differencesappeared to be along the dorsal surface. Note that the reg-istration of the thalami in this instance was solely for vis-ualization purposes and was not incorporated into theanalysis.DiscussionIt is well known that changes in the thalamus can be seenchronically after cortical injury [25,26]. This is not onlydue to direct effects of axonal damage, as axonal-sparingcortical lesions also result in thalamic degeneration [27].Such thalamic degeneration probably involves both anter-ograde and retrograde processes [28] and may be miti-gated by growth factors [29,30]. Brain development has acritical role on the extent of thalamic changes after a cor-tical lesion. Animal models have determined that perina-tal lesions are far less likely to induce thalamic changes,compared to when the cortical lesions are made prenatally[31] or in adulthood [32]. In contrast to the secondaryeffects of thalamic changes from cortical lesions, the tha-lamic changes in PD are related to selective non-dopamin-ergic neurodegeneration [1].Consistent with prior results, we found no significant dif-ferences in the volume of the thalami between PD subjectsand controls [3]. However, for the first time, we have dem-onstrated that the shape of the thalami undergoes system-atic changes in PD. The reason that shape may change butnot the volume may be due to the fact that particularnuclei (e.g CM-Pf) are involved, and thus, at the typicalresolution of MRI, this does not result in significantchanges in the overall volume. Alternately, since thalamicvolume may actually increase as a compensatory mecha-nism [12], other regions of the thalamus may hypertro-phy.The ability to non-invasively quantify subtle morphologi-cal shape changes appears to be a powerful technique. Weutilized standard structural MR imaging techniques with-out any special sequences nor any special scanner resolu-tion requirements. We obtained robust results frompooling data from two different centres using two differ-ent types of scanners.We used manual segmentation of the thalami in thispaper. Automatic segmentation of subcortical structures isan area of ongoing research [33], and often requires thetuning of many parameters, especially when the imagesmay be pooled from scanners from different centres. Sincethe person at each centre doing the segmentation wasblinded to disease status, it would be unexpected that aThe SPHARM-based method for shape determinationFigure 1The SPHARM-based method for shape determina-tion. The shape to be specified (the thalamus) and two con-centric spherical shells are shown. On the right is the intersection between the thalamus and shells as a function of rotation (θ) and azimuth (h). The rotation angle spans from 0 to 2π radians, and the azimuth angle is from 0 to π radians.Table 1: Results of volumetric and shape analysis. Numbers indicate the p-values obtained from the permutation test.Group Volume SPHARMControl, Left vs. Right 0.5630 0.1470PD, Left vs. Right 0.5780 0.0060Left thalamus, PD vs. Control 0.4150 0.0270Page 3 of 8(page number not for citation purposes)systematic bias was introduced into our results. Even then,any misspecification of the same ROI across subjectsRight thalamus, PD vs. Control 0.1730 0.0290BMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8Page 4 of 8(page number not for citation purposes)Two sets typical thalami registered and shown on brain from a PD subjectFigur  2Two sets typical thalami registered and shown on brain from a PD subject. Note that the registration here is solely for visualization purposes, and is not required for the calculation of shape differences. Also, although the thalami here were first smoothed with a 12 mm FWHM Gaussian kernel for visualization purposes, no smoothing was performed for the shape analysis and group comparison.BMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8would tend to increase inter-subject variability and pre-sumably reduce discriminability across groups making thetask harder for the shape analysis approach.We used SPHARM-based invariant descriptors to quantifythe shape of the thalami. The main advantages of such amethod is that it does not require that all brain images bewarped to a common space, nor does it require that thebrain images be aligned in any way. A drawback of theseinvariant features approach is that it is difficult to invertthe feature vectors, i.e. once given all the values of the dif-ferent invariants, it is impossible to reconstruct the origi-nal image which gave that feature vector – analogous tothe fact that given the volume of an object, it is impossibleto uniquely reconstruct the original shape. We thereforecannot create a "typical" brain structure by averaging thefeature vectors and create the image that would give thisfeature vector (e.g. to create an "average left thalamus").However, it is possible to cluster structures in the featurespace and find the brain structure whose feature vector isin the middle of the cluster so as to use it as an exemplarshape, which we have done (Figure 2).It is difficult to determine if the shape differences wedetected are attributable to any specific nuclei. However,based on prior pathological studies, it would be likely thatthe differences we detected were related to degenerationin the CM-Pf complex. Given that progressive supranu-clear palsy (PSP) has even greater involvement of VLpthan PD [4], it remains to be seen if thalamic shape is adiscriminable feature between these two conditions.We did not detect any association between overall shapechange and handedness, and dominant side presentationsor presence/absence of tremor. This may be due to the rel-atively small sample size employed in this study. How-ever, because the feature vectors consist of many differentcomponents, we don't discount that there may be a subsetof components that are sensitive to these disease parame-ters.ConclusionOur results suggest that systematic changes in thalamicshape can be non-invasively assessed in PD in vivo andthat shape changes, in addition to volume changes, mayrepresent a new avenue to assess the progress of neurode-generative processes. Although we cannot state whichparts of the thalamus are directly affected, previous path-ological studies would suggest that the shape changesdetected in this study represent degeneration in the centremedian-parafascicular (CM-Pf) complex, an area knownto represent selective non-dopaminergic degeneration inPD.MethodsThe study was approved by the appropriate InstitutionalReview Boards and Ethics Boards of the University of Brit-ish Columbia (UBC) and the University of North Caro-lina (UNC). All structural data were obtained as part offMRI studies whose results are reported elsewhere (e.g.,[34]).MR Imaging at the University of British ColumbiaAll subjects gave written informed consent prior to partic-ipating. Nine volunteers with clinically diagnosed PD par-ticipated in the study (5 men, 4 women, mean age 68.1 ±6.8 years, 7 right-handed, 2 left-handed). All subjects hadmild to moderate PD (Hoehn and Yahr stage 2–3) [35]with mean symptom duration of 3.6 ± 2.6 years. Werecruited ten healthy, age-matched control subjects with-out active neurological disorders (3 men, 7 women, meanage 55 ± 12.4 years, 9 right-handed, 1 left-handed). Exclu-sion criteria included atypical Parkinsonism, presence ofother neurological or psychiatric conditions and use ofantidepressants, sleeping tablets, or dopamine blockingagents.MRI was conducted on a Philips Achieva 3.0 T scanner(Philips, Best, The Netherlands) equipped with a 6 chan-nel Sense head-coil. A high resolution, three dimensional(3D) SPGR image data set of the whole brain consisting of170 axial slices at a FOV of 256 × 200 mm2 was acquiredfor WM/GM segmentation purposes and as anatomicalreference (inversion prepared 3D T1TFE, TR = 7.746 ms,TE = 3.55 ms, inversion delay = 880 ms, flip angle = 8.00°,voxel dimensions 1.0 × 1.0 × 1.0 mm3).The thalami were one of eighteen specific regions of inter-est (ROIs) that were manually drawn on each unwarped,aligned structural scan using the Amira software (MercuryComputer Systems, San Diego, USA). Although the tha-lami were manually segmented on the axial slices, theywere carefully examined in the coronal and saggital planesto ensure accuracy. The trained technician performing thesegmentation was blinded to the disease state.MR Imaging at the University of North CarolinaAll subjects gave written informed consent prior to partic-ipating. Nine volunteers with clinically diagnosed mild tomoderate PD (Hoehn and Yahr stage 2–3 – mean symp-tom duration of 2.1 ± 2.0 years) participated in the study(5 men, 4 women, mean age 58 ± 12 yrs, all right-handed). We also recruited eight healthy, age-matchedcontrol subjects without active neurological disorders (5men, 3 women, mean age 49 ± 14 yrs, 8 right-handed).Images were acquired on a 3.0 Tesla Siemens scanner (Sie-mens, Erlangen, Germany) with a birdcage-type standardPage 5 of 8(page number not for citation purposes)quadrature head coil and an advanced nuclear magneticresonance echoplanar system. The head was positionedBMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8along the canthomeatal line. Foam padding was used tolimit head motion. High-resolution T1 weighted anatom-ical images were acquired (3D SPGR, TR = 14 ms, TE = 7.7ms, flip angle = 25°, voxel dimensions 1.0 × 1.0 × 1.0 mm,176 × 256 voxels, 160 slices).ROIs (including thalami) were drawn manually by thesame trained research associate with assistance from mul-tiple on-line and published atlases (e.g. [36]).Thalami Shape AnalysisAs described in the technical appendix, the analysis ofeach shape results in a unique feature vector, of length n =1440. The left and right thalami were analyzed separately.For comparison, we examined for any differences in vol-ume. The volume of each thalamus was estimated as thenumber of voxels that each ROI contained multiplied bythe volume of a single voxel.To assess the significance of group differences betweenfeature vectors, we used a permutation test to generate anull distribution of Euclidean distances between featurevectors. The permutation test does not require a prioriassumptions about the data distribution, and is thus pre-ferred over T-test and F-test [37]. We assessed the differ-ences in left vs. right thalami in controls, left vs. right inPD subjects, PD vs. controls for the left thalamus, and PDvs. controls for the right thalamus. Although the bounda-ries of the thalami were determined by visual inspection,in prior work we compared feature vectors derived fromthalami segmented from structural scans obtained beforeand after giving L-dopa medication (as part of anotherfMRI study) [38]. As expected, no significant differencescould be detected in the two groups, suggesting that inde-pendent manual segmentation did not incur significantsystematic errors.Shape Analysis – technical aspectsLet Ψ(θ, φ) be a function defined on the unit spherewithθ and φ as the zenithal and azimuthal angles, respectively.The SPHARM representation for this function is given by(1) where  is the complex conjugate of the mthorder spherical harmonic of degree l. l ranges from 0 to L[16]. Increasing the value of L, also called the bandwidth,improves the representation accuracy at the cost of highercomputation time. This definition can also be extended toreal valued 3D distributions Ψ(r, θ, φ) (2), where r is thedistance from the origin to a given voxel. k is an indexintroduced to account for possible degeneracy due to theadditional dimension [39].As explained later in this section, rotationally invariantfeatures can be derived from this spherical harmonic rep-resentation. In our application, we also need the featuresto be invariant to any translation of the entire ROI in 3Dspace. To achieve this, we move the origin of the functionΨ(r, θ, φ), to the centroid of Ψs(r, θ, φ), where Ψs(r, θ, φ)is given by (3).Since direct computation of (2) is highly inefficient [40],we use an alternate approach by representing the data as aset of spherical functions obtained by intersecting the 3Ddata with spherical shells. Alternatively, for each value ofr, Ψ(r, θ, φ) can be visualized as a spherical shell compris-ing the function values at a distance r from the origin. rcan then be incremented in steps of t to encompass theentire ROI. If the initial representation of the function isin the form of a cubic grid (regular isotropic voxels in ourcase), volumetric interpolation is required to resample theROI in the spherical coordinate space.When analyzing multiple subjects' ROIs simultaneously,we define the maximum radius, Rmax, as the minimumradial distance in voxel count that encompasses all non-zero values of all subjects' ROIs being analyzed. To repre-sent the values from the cubic grid of all ROIs with suffi-cient accuracy, 2Rmax shells are used. To achieve scaleinvariance, the shells must be distributed evenly through-out the spatial extent of each ROI. Since the ROI sizeacross subjects is not uniform, shell spacing t must beadjusted for each subject separately. This procedureensures that each shell captures similar features from the3D ROI irrespective of its scale.Surface sampling along each of these shells is performedon an equiangular spherical grid of dimensions 2L × 2L[40]. The common bandwidth L for all shells of all func-tions is chosen to satisfy the sampling criterion for thelargest shell in this set of ROI, namely the one with radiusRmax. The surface area for this shell represents the maxi-Ylm∗ ( )θ φ,c d Y dlmlm= ∫ ∫ ∗φ θ φ θ φ θ θπ π020( , ) ( , )sin( )Ψ (1)c r dr dkrrY r dklmlm=∞∗∫ ∫∫20 0022φ π θ φ θ φ θ θππsin( )( , ) ( , , )sin( )Ψ(2)ΨΨΨsrif rif r, ,, ,, ,θ φθ φθ φ( ) = ( ) ≠( ) =⎧⎨⎪⎩⎪1 00 0(3)Page 6 of 8(page number not for citation purposes)mum surface shell area that needs to be sampled by theequiangular grid; hence, any value of L satisfying theBMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8required equiangular sampling (2L × 2L) at this shell willbe sufficient to represent data from smaller radii shells.The minimum value for L is obtained by equating the sur-face area of this largest shell to the equiangular samplinggrid (4). Higher values of L are not used, since it increasecomputation time with no added benefit. Also, this willresult in longer feature vectors, complicating the analysis.Furthermore, when the represented object is a discretearray, higher values of l, resulting from a larger L, may cor-respond to sampling noise [39]. Recognizing that inapplications pertaining to discrimination, high accuracyin the SPHARM representation is not a necessity, we choseto use the minimum value for L as that obtained by (4).To obtain the SPHARM representation for all shells, a dis-crete SPHARM transform is performed at each value of r toobtain  (5). Features derived from this representation,however, do not provide a unique function representation[41,42]. For instance, rotating the inner and outer shellsby different amount will result in different spatial distri-bution of function values. However, in this approach, thederived features are insensitive to these rotational trans-forms, thus resulting in the same feature values for dissim-ilar spatial distributions.Burel and Henocq's original equation (2) does not havethis problem, since a part of the basis function is a func-tion of r. However, since (2) is computationally intracta-ble [17], we proposed an efficient approach that uses aradial transform (6), derived from (2), to obtain a uniquefunction representation. The transform (6) retains the rel-ative orientation information of the shells, thus the fea-tures derived will be sensitive to independent rotations ofthe different shells, thereby ensuring that unique featurerepresentation is obtained.The range of k could be changed to obtain differentlengths of the final feature vector. However, to avoidunnecessarily increasing the feature vector length or los-range, we choose to keep the range of k the same as thatof r, i.e. 2Rmax.From the obtained representation (6), we then computesimilarity transform invariant features using (7) for eachvalue of l and k [39] with p and q are used to index thesefeatures. Note we reshape I into a single row vector ofdimensions D = L × 2Rmax for later analysis.In order to provide a scalar estimate how different a giventhalamus shape was, we calculated the mean of the rowvectors, I (Eqn 7) separately for both the left and right tha-lami. A distance metric, estimating how "abnormal" agiven shape was estimated by determining the Euclideandistance between the given feature vector and the meanvector. For example, the distance for right thalamus for thejth subject was estimated as:Competing interestsThe author(s) declare that they have no competing inter-ests.Authors' contributionsMJM conceptualized the study, supervised the collectionof the data from UBC, and supervised the application ofthe SPHARM technique to medical data. AU developedthe SPHARM-based method and performed the calcula-tions. RA supervised the development of the SPHARM-based method and assisted in the application to medicaldata. SP collected the data at UBC and performed themanual segmentations of the thalami. ML collected thedata at UNC and manually segmented the data fromUNC. XH supervised the collection of UNC data andassisted in biological interpretation of the results.AcknowledgementsThis work was supported by a grant from NSERC/CIHR CHRP-(323602 – 06) (MJM).References1. Bacci JJ, Kachidian P, Kerkerian-Le Goff L, Salin P: Intralaminar tha-lamic nuclei lesions: widespread impact on dopamine dener-vation-mediated cellular defects in the rat basal ganglia.  JNeuropathol Exp Neurol 2004, 63(1):20-31.2. Henderson JM, Carpenter K, Cartwright H, Halliday GM: Degener-ation of the centre median-parafascicular complex in Parkin-4 2 22π πR L L L Rmax max,= × = (4)c rlmc d Y r dr Rrlmlm==∫ ∫ ∗φ θ φ θ φ θ θπ π0201 2 3 2( , ) ( , , )sin( )[ , , ,..., ]maxΨ(5)c rkrrck RklmrlmrR===∑ 21221 2 3 2sin( )[ , , ,..., ]maxmaxπ(6)I p q c c p L q l Rklmklmm lm lk lk R, *, ... , ...maxmax( ) = ( ) = ==−===∑∑21 2(7)Dist I IrightjkjkDrightj= −=∑( )12 (8)Page 7 of 8(page number not for citation purposes)ing any important information caused by reducing the son's disease.  Ann Neurol 2000, 47(3):345-352.3. Henderson JM, Carpenter K, Cartwright H, Halliday GM: Loss ofthalamic intralaminar nuclei in progressive supranuclearBMC Neurology 2008, 8:8 http://www.biomedcentral.com/1471-2377/8/8palsy and Parkinson's disease: clinical and therapeutic impli-cations.  Brain 2000, 123(Pt 7):1410-1421.4. Halliday GM, Macdonald V, Henderson JM: A comparison ofdegeneration in motor thalamus and cortex between pro-gressive supranuclear palsy and Parkinson's disease.  Brain2005, 128(Pt 10):2272-2280.5. Aymerich MS, Barroso-Chinea P, Perez-Manso M, Munoz-Patino AM,Moreno-Igoa M, Gonzalez-Hernandez T, Lanciego JL: Conse-quences of unilateral nigrostriatal denervation on the thala-mostriatal pathway in rats.  Eur J Neurosci 2006, 23(8):2099-2108.6. Kerkerian-Le Goff L, Bacci JJ, Salin P, Aymerich MS, Barroso-ChineaP, Obeso JA, Lanciego JL: Intralaminar Thalamic Nuclei areMain Regulators of Basal Ganglia: Possible involvement inthe pathophysiology of Parkinson’s disease.  In The Basal GangliaVIII Edited by: Bolam JP, Ingham CA, Magill PJ.  Springer US;2005:331-339. 7. Nakamura Y, Otake K, Tokuno H: The parafascicular nucleusrelays spinal inputs to the striatum: an electron microscopestudy in the rat.  Neurosci Res 2006, 56(1):73-79.8. Lacey CJ, Bolam JP, Magill PJ: Novel and distinct operational prin-ciples of intralaminar thalamic neurons and their striatalprojections.  J Neurosci 2007, 27(16):4374-4384.9. Kimura M, Minamimoto T, Matsumoto N, Hori Y: Monitoring andswitching of cortico-basal ganglia loop functions by the tha-lamo-striatal system.  Neurosci Res 2004, 48(4):355-360.10. Ogawa T, Yoshida Y, Okudera T, Noguchi K, Kado H, Uemura K:Secondary thalamic degeneration after cerebral infarction inthe middle cerebral artery distribution: evaluation with MRimaging.  Radiology 1997, 204(1):255-262.11. Tamura A, Tahira Y, Nagashima H, Kirino T, Gotoh O, Hojo S, SanoK: Thalamic atrophy following cerebral infarction in the ter-ritory of the middle cerebral artery.  Stroke 1991,22(5):615-618.12. Hulshoff Pol HE, van der Flier WM, Schnack HG, Tulleken CA, RamosLM, van Ree JM, Kahn RS: Frontal lobe damage and thalamicvolume changes.  Neuroreport 2000, 11(13):3039-3041.13. Kassubek J, Juengling FD, Hellwig B, Spreer J, Lucking CH: Thalamicgray matter changes in unilateral Parkinsonian restingtremor: a voxel-based morphometric analysis of 3-dimen-sional magnetic resonance imaging.  Neurosci Lett 2002,323(1):29-32.14. Summerfield C, Junque C, Tolosa E, Salgado-Pineda P, Gomez-AnsonB, Marti MJ, Pastor P, Ramirez-Ruiz B, Mercader J: Structural brainchanges in Parkinson disease with dementia: a voxel-basedmorphometry study.  Arch Neurol 2005, 62(2):281-285.15. Kaasinen V, Maguire RP, Hundemer HP, Leenders KL: Corticostri-atal covariance patterns of 6-[18F]fluoro-L-dopa and[18F]fluorodeoxyglucose PET in Parkinson's disease.  J Neurol2006, 253(3):340-348.16. Bookstein FL: "Voxel-based morphometry" should not beused with imperfectly registered images.[see com-ment][comment].  Neuroimage 2001, 14(6):1454-1462.17. Ashburner J, Friston KJ: Why voxel-based morphometry shouldbe used.[comment].  Neuroimage 2001, 14(6):1238-1243.18. Miller MI, Beg MF, Ceritoglu C, Stark C: Increasing the power offunctional maps of the medial temporal lobe by using largedeformation diffeomorphic metric mapping.  Proc Natl Acad SciU S A 2005, 102(27):9685-9690.19. Stark CEL, Okado Y: Making memories without trying: medialtemporal lobe activity associated with incidental memoryformation during recognition.  J Neurosci 2003,23(17):6748-6753.20. Yushkevich PA, Detre JA, Mechanic-Hamilton D, Fernandez-SearaMA, Tang KZ, Hoang A, Korczykowski M, Zhang H, Gee JC: Hippoc-ampus-specific fMRI group activation analysis using the con-tinuous medial representation.  Neuroimage 2007,35(4):1516-1530.21. McKeown MJ, Li J, Huang X, Wang ZJ: Local Linear DiscriminantAnalysis (LLDA) For Inference Of Multisubject fMRI Data. :Honolulu, Hawaii.   Volume 1. ; 2007:305 -3308. 22. Mangin JF, Poupon F, Duchesnay E, Riviere D, Cachia A, Collins DL,Evans AC, Regis J: Brain morphometry using 3D momentinvariants.  Med Image Anal 2004, 8(3):187-196.23. Ng B, Abu-Gharbieh R, Huang X, McKeown MJ: Characterizing24. Uthama A, Abugharbieh R, Traboulsee A, McKeown MJ: InvariantSPHARM Shape Descriptors for Complex Geometry in MRRegion of Interest Analysis: Lyon, France.   ; 2007. 25. Matthews MA: Death of the central neuron: an electron micro-scopic study of thalamic retrograde degeneration followingcortical ablation.  Journal of Neurocytology 1973, 2(3):265-288.26. Vicedomini JP, Corwin JV, Nonneman AJ: Role of residual anteriorneocortex in recovery from neonatal prefrontal lesions inthe rat.  Physiol Behav 1982, 28(5):797-806.27. Ross DT, Ebner FF: Thalamic retrograde degeneration follow-ing cortical injury: an excitotoxic process?  Neuroscience 1990,35(3):525-550.28. Sorensen JC, Dalmau I, Zimmer J, Finsen B: Microglial reactions toretrograde degeneration of tracer-identified thalamic neu-rons after frontal sensorimotor cortex lesions in adult rats.Experimental Brain Research 1996, 112(2):203-212.29. Kumon Y, Sakaki S, Watanabe H, Nakano K, Ohta S, Matsuda S,Yoshimura H, Sakanaka M: Ciliary neurotrophic factor attenu-ates spatial cognition impairment, cortical infarction andthalamic degeneration in spontaneously hypertensive ratswith focal cerebral ischemia.  Neuroscience Letters 1996, 206(2-3):141-144.30. Yamada K, Kinoshita A, Kohmura E, Sakaguchi T, Taguchi J, KataokaK, Hayakawa T: Basic fibroblast growth factor prevents tha-lamic degeneration after cortical infarction.  J Cereb Blood FlowMetab 1991, 11(3):472-478.31. Jackson GF 3rd, Villablanca JR, Loopuijt LD: Few neocortial andthalamic morphological changes after a neonatal frontal cor-tical ablation contrast with the effects of a similar lesion infetal cats.  Brain Research 1995, 90(1-2):62-72.32. Villablanca JR, Burgess JW, Benedetti F: There is less thalamicdegeneration in neonatal-lesioned than in adult-lesioned catsafter cerebral hemispherectomy.  Brain Res 1986,368(2):211-225.33. Amini L, Soltanian-Zadeh H, Lucas C, Gity M: Automatic segmen-tation of thalamus from brain MRI integrating fuzzy cluster-ing and dynamic contours.  IEEE Trans Biomed Eng 2004,51(5):800-811.34. Lewis MM, Slagle CG, Smith AB, Truong Y, Bai P, McKeown MJ, Mail-man RB, Belger A, Huang X: Task specific influences of Parkin-son's disease on the striato-thalamo-cortical and cerebello-thalamo-cortical motor circuitries.  Neuroscience Letters 2007,147(1):224-235.35. Hoehn MM, Yahr MD: Parkinsonism: onset, progression andmortality.  Neurology 1967, 17(5):427-442.36. Damasio H: Human Brain Anatomy in Computerized Images.2005.37. Ludbrook J: Advantages of permutation (randomization) testsin clinical and experimental pharmacology and physiology.Clin Exp Pharmacol Physiol 1994, 21(9):673-686.38. Tootoonian S, Abugharbieh R, Huang X, McKeown MJ: Shape vs.Volume: Invariant Shape Descriptors for 3D Region of Inter-est Characterization in MRI.  In IEEE International Symposium onBiomedical Imaging (ISBI) Arlington-USA ; 2006. 39. Burel G, Henocq H: Three-dimensional invariants and theirapplication to object recognition.  Signal Process 1995,45(1):1-22.40. Healey DM Jr., Rockmore DN, Moore SB: An FFT for the 2-sphere and applications.  1996, 3:1323-1326 vol. 3.41. Vranic DV: An improvement of rotation invariant 3D shapedescriptor based on functions on concentric spheres: Barce-lona.   Volume 3. ; 2003:757-760. 42. Kazhdan M, Funkhouser T, Rusinkiewicz S: Rotation invariantspherical harmonic representation of 3D shape descriptors.In Proceedings of the 2003 Eurographics/ACM SIGGRAPH symposium onGeometry processing Aachen, Germany , Eurographics Association;2003. Pre-publication historyThe pre-publication history for this paper can be accessedhere:Page 8 of 8(page number not for citation purposes)fMRI Activations within Regions of Interest (ROIs) Using 3DMoment Invariants: New York.   ;  2006:63. http://www.biomedcentral.com/1471-2377/8/8/prepub

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0167789/manifest

Comment

Related Items