Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print January 14, 2008, 10.1148/radiol.2463070486
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2463070486v1
246/3/880    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Poonawalla, A. H.
Right arrow Articles by Narayana, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Poonawalla, A. H.
Right arrow Articles by Narayana, P. A.
(Radiology 2008;246:880-886.)
© RSNA, 2008

Neuroradiology

Diffusion-Tensor MR Imaging of Cortical Lesions in Multiple Sclerosis: Initial Findings1

Aziz H. Poonawalla, PhD, Khader M. Hasan, PhD, Rakesh K. Gupta, MD, Chul W. Ahn, PhD, Flavia Nelson, MD, Jerry S. Wolinsky, MD, and Ponnada A. Narayana, PhD

1 From the Departments of Diagnostic and Interventional Radiology (A.H.P., K.M.H., P.A.N.), Internal Medicine (C.W.A.), and Neurology (F.N., J.S.W.), the University of Texas Medical School at Houston, 6431 Fannin St, Houston, TX 77030; and Department of Radiodiagnosis, King George's Medical University, Lucknow, India (R.K.G.). Received March 14, 2007; revision requested May 23; revision received June 12; accepted July 18; final version accepted September 14. Supported by National Institutes of Health grants R01 EB02095 (P.A.N.), S10 RR19186 (P.A.N.), and R01 NS052505 (K.M.H.). Address correspondence to P.A.N. (e-mail: ponnada.a.narayana{at}uth.tmc.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively perform a direct measurement of fractional anisotropy (FA) and mean diffusivity (MD) in cortical lesions of patients with multiple sclerosis (MS).

Materials and Methods: The study was approved by the institutional review board and was HIPAA compliant; informed consent was obtained. Magnetic resonance (MR) images, including double inversion-recovery (DIR), phase-sensitive inversion-recovery (PSIR), and diffusion-tensor images, were acquired from nine MS patients with cortical lesions (five women, four men; median age, 47 years) and nine age- and sex-matched volunteer control subjects. Following nonlinear elastically constrained image registration for aligning diffusion-weighted images to DIR images, maps of FA and MD were computed for each subject. Cortical lesions were identified on DIR images and validated by using PSIR images. The diffusion-tensor imaging maps were then overlaid on the coregistered DIR images, and mean FA and MD values were measured in regions of interest drawn on the cortical lesions. Differences between normal gray matter (GM) and cortical lesions were evaluated by using the generalized estimating equation. FA and MD histograms of whole brain and GM (global analysis) in healthy control subjects and MS patients were also computed for comparison with those in previously published studies.

Results: FA and MD values were significantly higher in cortical lesions compared with similar regions in healthy control subjects. Histogram peak FA was significantly decreased and peak MD was significantly increased in patients relative to control subjects.

Conclusion: DIR and PSIR combined with nonlinear image registration allowed direct focal measurement of FA and MD in cortical lesions.

© RSNA, 2008


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Diffusion-tensor magnetic resonance (MR) imaging of cortical lesions in multiple sclerosis (MS) has the potential to provide important insights into their microstructural environment and underlying pathologic changes. Although histopathologic studies suggest the presence of numerous cortical lesions in MS (1–3), reliable detection of these lesions by using conventional MR imaging is challenging (2,4,5). Therefore, to date, all studies applying diffusion-tensor imaging to the cortex have focused on normal-appearing gray matter (GM). For example, results of global histogram analyses of the whole brain suggest increased mean diffusivity (MD) and decreased fractional anisotropy (FA) in MS patients relative to healthy control subjects (6–8). Results of histogram analyses of the normal-appearing GM in patients have also indicated increased MD and/or decreased FA compared with healthy control subjects (8–11). In addition to results with these global histogram-based approaches, results with focal region of interest (ROI) analysis of cortical normal-appearing GM have also been reported, but they are inconclusive (10,12).

Although the bulk analyses of normal-appearing GM are important, a direct probe of the microstructural environment of cortical lesions could be more informative in regard to the underlying local pathologic changes. A prerequisite for such analyses is more reliable and accurate identification of cortical lesions with advanced MR imaging techniques, such as double inversion recovery (DIR), in which both cerebrospinal fluid (CSF) and white matter (WM) are selectively suppressed (13–15). DIR MR imaging was shown to be superior to conventional MR imaging in the detection of cortical lesions (16,17). However, even with recent advances in high-field-strength MR imaging, DIR MR images are inherently noisy, are susceptible to flow-related artifacts (14), and have regional variations in GM signal intensity (17), all of which can result in false-positive findings for lesions (18). The reliability of cortical lesion identification can be greatly improved by combining DIR with another sequence that better delineates the GM-WM junction, such as T1-weighted phase-sensitive inversion recovery (PSIR) (18,19).

Most diffusion-tensor imaging studies employ echo-planar imaging, which is susceptible to geometric distortions induced by eddy current (20). Even with correction methods for eddy current, substantial distortions and signal intensity loss can persist in the echo-planar imaging data, particularly at the level of the sinuses and frontal lobes. Therefore, focal diffusion-tensor imaging analysis of cortical lesions also requires accurate coregistration of diffusion-weighted images to the images used for lesion identification. Standard linear registration methods are inadequate for coregistering diffusion-tensor imaging data to fast spin-echo–based sequences, including DIR. Advanced nonlinear image registration methods, such as those for which elastic deformation models are employed (21), have the potential to greatly alleviate this problem.

The purpose of our study was to prospectively perform a direct measurement of FA and MD in cortical lesions of patients with MS.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Participants
Our study was approved by the Institutional Review Board of the University of Texas Health Science Center at Houston, Houston, Tex, and was Health Insurance Portability and Accountability Act compliant. Written informed consent was obtained from all participants. Nine patients with relapsing-remitting MS (five women, four men) and a median age of 47 years (range, 22–55 years) were selected from a larger pool of 149 patients imaged during a period of 22 months from September 2004 to July 2006. The patients were selected on the basis of retrospective identification of cortical lesions on previous MR images, as described elsewhere (18). These nine patients were the only ones who met this inclusion criterion. Nine healthy volunteers (five women, four men) with a median age of 47 years (range, 23–52 years) and no known brain abnormalities or neurologic disorders were recruited.

Image Acquisition
All patients and control volunteers were imaged with a 3-T MR unit (Intera; Philips Medical Systems, Best, Netherlands) that had a gradient system (Quasar; Philips Medical Systems). The maximum gradient amplitude was 80 mT/m, and the slew rate was 200 mT/m/sec. A six-element head coil (Philips Medical Systems) was used. The MR imaging protocol included dual-echo fast spin-echo, fluid-attenuated inversion-recovery, single-shot echo-planar diffusion-tensor, WM- and CSF-suppressed DIR, and PSIR reconstruction sequences (19). For each sequence, 44 contiguous and interleaved transverse sections of 3 mm thickness were acquired for full-brain coverage, with an image matrix of 256 x 256 and a 24-cm square field of view. The diffusion-tensor imaging acquisition matrix was 112 x 112, with homodyne reconstruction, and was interpolated to 256 x 256. With the diffusion-tensor imaging protocol, an icosahedral-encoding scheme with 21 directions was employed. For all sequences, a sensitivity encoding acceleration factor of 2.0 was employed (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Acquisition Parameters for MR Imaging Sequences

 
For quality control, identical diffusion-tensor imaging acquisitions were performed with a spherical water phantom. Because of the isotropic nature of water diffusion, the expected value of FA is zero, so any residual anisotropy can be solely attributable to noise and/or bias introduced by the encoding scheme. The mean FA and standard deviation were calculated throughout the whole volume of the phantom by using the complete set of 21 directions, also called "icosa21," as well as an icosahedral subset of six directions, also called "icosa6," for evaluation of the relative bias, as described by Madi et al (22). These measurements were repeated seven times during a 6-month period to assess the reproducibility and stability of the diffusion-tensor imaging results.

Postprocessing
All image postprocessing was performed by a medical physicist (A.H.P., with 8 years of experience). Diffusion-weighted images were intraregistered to the baseline images obtained with a b value of 0 sec/mm2 to correct for the distortions caused by eddy current by using a workstation (Pride; Philips Medical Systems). Next, the baseline images were interregistered to the fast spin-echo images by using a viscoelastic nonlinear registration method that was based on a regionally varying elastic deformation (21). FA and MD maps were then computed from the postregistered diffusion-tensor imaging data.

For each subject, a linear registration model was used to align the fast spin-echo–based images (fluid-attenuated inversion-recovery, PSIR, and DIR images) to the dual-echo fast spin-echo space, and image segmentation was then performed by using an automated hybrid parametric-nonparametric algorithm (23,24). Patient data were segmented into WM, GM, CSF, T2 hyperintense lesions (23), and black holes (24). In control volunteers, images were segmented into WM, GM, and CSF only.

Lesion Identification
All patient images were interpreted with consensus by two experienced neurologists with clinical and imaging experience in MS (J.S.W. and F.N., with 17 and 6 years of experience, respectively), a neuroradiologist (R.K.G., with 15 years of experience), and a medical physicist specializing in MR imaging (P.A.N., with 20 years of experience). A custom-made software tool developed in house at our laboratory was used for visualizing the coregistered DIR and PSIR images side by side with cursor tracking so that landmarks on one image could be immediately associated with the corresponding anatomy on the other. For each patient, cortical lesions were first identified on DIR images as localized hyperintense regions. Those regions with no hypointense analogue on PSIR images were discarded as false-positive findings (Fig 1) (18). At inspection of the GM-WM boundary on PSIR images, lesions with a volume that extended more than approximately 25% into the subcortical WM were excluded from the analysis. Lesions with questionable WM involvement from poor lesion border definition were also excluded.


Figure 1A
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Transverse MR images in MS patient acquired with (a) DIR (15 000/32/325, 3400) and (b) PSIR (4300/8/400) sequences. Cortical lesion (arrow) appears as hyperintense region on a and as corresponding hypointense region on b.

 

Figure 1B
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Transverse MR images in MS patient acquired with (a) DIR (15 000/32/325, 3400) and (b) PSIR (4300/8/400) sequences. Cortical lesion (arrow) appears as hyperintense region on a and as corresponding hypointense region on b.

 
Focal Analysis
ROI analysis was performed by the same medical physicist specializing in MR imaging who performed image postprocessing and the same neuroradiologist who performed lesion identification. Another custom-made software tool also developed in house was used for ROI analysis, which enabled triple-plane reformatted views of color-coded FA and MD maps to be superimposed on the DIR images (Fig 2). For each patient, the FA map was overlaid with a threshold value of 0.2 so that the major WM fiber tracts were clearly visible. As a result of blurring from interpolation on diffusion-tensor images, some lesion candidates were partially occluded by a major WM tract; these lesions were excluded from analysis. For each remaining lesion, an ROI of 3–6 voxels, depending on lesion size, was carefully placed so as to maximize the DIR signal intensity and minimize MD to avoid potential CSF contamination. Mean FA and MD values for the lesion were then recorded, along with the standard deviation within the ROI. For healthy control subjects, ROIs were drawn in the cortical GM in the frontal, medial, occipital, parietal, and temporal lobes. ROIs were drawn on both left and right sides that were completely within the cortical ribbon but well away from the major WM tracts.


Figure 2
View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Transverse color-coded FA map (repetition time msec/echo time msec, 7100/65; b = 1000 sec/mm2; icosahedral directions, 21) superimposed on DIR image (15 000/32/325, 3400) in MS patient. Cortical lesion on DIR image is observed (arrow). FA threshold value was 0.2. On the red-green-blue FA map, red = right to left directions, green = anterior to posterior directions, and blue = superior to inferior directions.

 
Prior to statistical analysis, all ROI measurements were filtered to minimize possible partial volume (PV) contamination from WM and CSF, as follows. For a GM voxel, the effect of WM contamination would be to increase FA, whereas the effect of CSF contamination would be to increase MD. For a voxel containing a fraction {alpha} of WM contamination and a 1 – {alpha} fraction of GM, we can define the FA and MD as a weighted sum of the two compartments thus: FAPV = {alpha} · FAWM + (1 – {alpha}) · FAGM and MDPV = {alpha} · MDCSF + (1 – {alpha}) · MDGM, where the subscripts PV, WM, and GM refer to the values of diffusion-tensor imaging metrics for voxels with PV and pure WM and GM, respectively. We employed a value of FAWM = 0.2 (the minimum value that is reported for WM in the published literature) and a value of MDCSF = 0.003 mm2/sec and took the maximum of our measurements in normal control cortical GM to obtain FAGM = 0.11 and MDGM = 0.0011 mm2/sec. With a conservative definition of {alpha} = 0.25, we found that FAPV = 0.1325 and MDPV = 0.0016 mm2/sec. Therefore, if the mean ROI measurements exceeded either of these limits, the entire ROI was discarded and excluded from the analysis.

The set of nine patients yielded an initial pool of 53 lesion candidates. Of these, 21 were excluded because of partial occlusion from a WM tract, and a further 11 were excluded for PV contamination, leaving 21 cortical lesions with clearly measurable diffusion-tensor imaging metrics. There were 90 ROI measurements in normal GM, none of which required exclusion for PV.

Global Analysis
Global histogram analysis was performed by the same medical physicist specializing in MR imaging who performed the image postprocessing and the ROI analysis. The output of the segmentation algorithm was used to generate tissue-specific binary masks for each patient and healthy subject. These masks were applied to the diffusion-tensor imaging maps for generating tissue-specific histograms of MD and FA. Whole-brain histograms also were computed. The first bin in the histograms was removed to minimize the background, and the histograms were normalized to the total voxel count. Peak values of FA and MD for each histogram were recorded without any filtering or smoothing beforehand.

Statistical Analysis
All statistical analyses were performed by a biostatistician (C.W.A., with 20 years of experience). All ROI data were grouped into two sets, one for patients and one for control subjects. The one-tailed generalized estimating equation was used to test whether control ROIs had significantly decreased mean FA and decreased mean MD values than did patient ROIs. This method accounts for the correlation of parameter values within a subject. For the histogram comparisons, the one-tailed Student t test was used to assess whether healthy volunteers had significantly increased peak FA and decreased peak MD values than did MS patients for each of the tissue compartments. For the water phantom experiments, analysis of variance was performed for evaluation of temporal stability. Differences with a P value of less than .05 were considered significant for all statistical tests. Analyses were performed with statistical software (SAS, version 9.1, 2006; SAS Institute, Cary, NC) and spreadsheet software (Microsoft Excel, version 11, 2003; Microsoft, Redmond, Wash).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
The mean FA values measured in the water phantom were 0.036 ± 0.001 and 0.059 ± 0.002 by using the icosa21 and icosa6 schemes, respectively. The bias of measurements with the use of icosa6 relative to icosa21 was highly significant (P < .001, analysis of variance), and was observed consistently with time. Results of repeated analysis of variance did not indicate any significant difference in the FA values measured at different times.

Focal Analysis
The diffusion-tensor imaging measurements, expressed as group mean ± standard deviation across all ROIs (Table 2), indicate that MD and FA values were significantly increased (P < .001) in cortical lesions in patients relative to cortical GM in healthy control subjects.


View this table:
[in this window]
[in a new window]

  		Table 2. Comparison of Diffusion-Tensor Imaging Metric Values for ROI Measurements in Cortical GM and Cortical Lesions

 
Global Analysis
On the basis of the whole-brain analysis, the peak FA value was significantly decreased (P = .038) for MS patients relative to healthy control subjects, whereas peak MD was significantly increased (P = .01). The same results were observed for the GM compartment alone (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 3. Comparison of Histogram Peak Values for MD and FA in GM Compartment Alone and Whole Brain in Healthy Volunteers and Patients

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
To our knowledge, ours is the first study in which diffusion-tensor imaging metrics were directly measured in cortical lesions in MS patients. The primary finding of our study was that mean FA values of cortical lesions in MS patients were significantly increased relative to the cortical GM in healthy control subjects. This behavior is opposite to that of MS lesions in WM, which exhibit decreased FA values that are indicative of disrupted microstructural organization (25). At first glance, the results of our study may appear to contradict those in recent reports in which decreased FA was observed in cortical normal-appearing GM, as measured by using bulk histogram analysis (10). However, results of our own global histogram analysis completely agree with those published findings. Global histogram analysis includes GM, normal-appearing GM, and focal cortical lesions. The volume of focal lesions is much smaller than the total volume of GM and normal-appearing GM. Therefore, any focal microstructural changes occurring in cortical lesions, as inferred by findings at diffusion-tensor imaging, are likely to be obscured or overwhelmed by results of global analysis of GM and normal-appearing GM. This underscores the importance of focal analysis.

Our results of increased anisotropy in cortical lesions may perhaps be understood in the context of the published histologic characteristics of cortical lesions. Cortical lesions have shown evidence of targeted transection of neurites and loss of dendritic arborization of cortical neurons (26), as well as generalized neuronal, synaptic, and glial loss (3). The loss of dendrites and axons is relevant to our findings, as they are known to contribute to diffusion anisotropy in normal GM (27). Reduced dendritic arborization could increase coherence, which may manifest as increased diffusion anisotropy. It is interesting to note that increased FA values, attributed to targeted dendrite degeneration, were also reported to have been observed in the caudate nuclei of patients with Huntington disease (28). The increased FA is also consistent with increased activation of microglia with extended processes that ensheathe neuronal cell bodies and neurites (26). It is conceivable that activated microglia would also contribute to increased local organization of the cortical microstructure, potentially increasing local diffusion anisotropy.

Another interesting point is that our measured FA values in normal cortical GM are significantly smaller than other published values (10,29). We measured a mean FA value of 0.084, compared with a mean FA value of 0.22 reported by Bhagat and Beaulieu (29) and a mean FA value of 0.28 reported by Vrenken et al (10). Researchers in both of these previous studies employed diffusion-encoding schemes with only six diffusion directions that were not rotationally invariant (30). An increased number of directions beyond six is preferable for both accurate and unbiased values, as demonstrated empirically by the results of our water phantom experiment and theoretically by the results in other studies (30,31). The icosa21 scheme used in our study has been shown to be robust in this regard (22,25). In addition, Vrenken et al (10) acknowledged that their use of a large section thickness (6 mm) was not optimal and probably led to WM contamination of their cortical GM measurements. These factors, along with our quality control results, suggest that the FA values we measured in the cortex are robust and accurate.

A methodological limitation of our study was that we did not employ CSF suppression for our diffusion-tensor imaging acquisition. As noted in Materials and Methods, ROI placement was guided by CSF concerns to some degree, and we also set a fairly conservative threshold level for PV contamination filtering on MD values. With respect to FA values, PV caused by CSF is expected to reduce FA (32); our findings of increased FA would be further strengthened by including CSF suppression into the diffusion-tensor imaging sequence in future studies.

The accuracy of the nonlinear registration of diffusion-tensor imaging data to the fast spin-echo–based sequences is another critical issue. Misregistration could lead to potential WM contamination of our GM ROI measurements, leading to a false apparent increase in FA. The nonlinear registration algorithm we employed in this study (21) substantially outperformed other algorithms, such as automated image registration (33), and such performance led to excellent agreement between the DIR images and diffusion-tensor imaging maps. We acknowledge that the issue of registration could largely be avoided with use of alternate diffusion-tensor imaging acquisition methods. For example, Vrenken et al (10) used a stimulated-echo acquisition mode–based diffusion pulse sequence to acquire their data with greatly reduced geometric distortions from eddy currents. However, stimulated-echo acquisition mode is associated with a significant signal-to-noise ratio penalty, lower spatial resolution, and increased acquisition time relative to fast sequences, such as the echo-planar imaging sequence. We believe that software-based approaches provide a more pragmatic solution overall, especially for quantitative studies that require an increased number of diffusion directions.

The primary limitation of our study was the small sample size of valid cortical lesion candidates. Though DIR substantially improved the detection rate of cortical lesions relative to conventional MR imaging, it is likely that most of the cortical lesions remained undetected. Of a fairly extensive patient population (n = 149), only 6% had enough lesions visible on DIR images and validated on PSIR images to justify inclusion in this study. The conservative approach to ROI placement and PV filtering further reduced the available pool of measurement samples.

In conclusion, direct focal analysis of cortical lesions with diffusion-tensor imaging reveals important characteristics that are masked by global histogram analysis. Reliable identification by using DIR and PSIR sequences combined with nonlinear image registration allowed direct focal measurement of diffusion-tensor imaging metrics in cortical lesions.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
We acknowledge Vipulkumar Patel, RT(MR), for invaluable assistance with image scanning and protocol optimization.


   FOOTNOTES
 

Abbreviations: CSF = cerebrospinal fluid • DIR = double inversion recovery • FA = fractional anisotropy • GM = gray matter • MD = mean diffusivity • MS = multiple sclerosis • PSIR = phase-sensitive inversion recovery • PV = partial volume • ROI = region of interest • WM = white matter

Author contributions: Guarantors of integrity of entire study, A.H.P., P.A.N.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, A.H.P., R.K.G., P.A.N.; clinical studies, A.H.P., K.M.H., F.N., J.S.W., P.A.N.; experimental studies, K.M.H.; statistical analysis, C.W.A.; and manuscript editing, A.H.P., K.M.H., R.K.G., F.N., J.S.W., P.A.N.

Authors stated no financial relationship to disclose.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Bö L, Geurts JJ, Mork SJ, van der Valk P. Grey matter pathology in multiple sclerosis. Acta Neurol Scand Suppl 2006;183:48–50.[Medline]
  2. Kidd D, Barkhof F, McConnell R, Algra PR, Allen IV, Revesz T. Cortical lesions in multiple sclerosis. Brain 1999;122:17–26.[Abstract/Free Full Text]
  3. Wegner C, Esiri MM, Chance SA, Palace J, Matthews PM. Neocortical neuronal, synaptic, and glial loss in multiple sclerosis. Neurology 2006;67:960–967.[Abstract/Free Full Text]
  4. Bakshi R, Ariyaratana S, Benedict RH, Jacobs L. Fluid-attenuated inversion recovery magnetic resonance imaging detects cortical and juxtacortical multiple sclerosis lesions. Arch Neurol 2001;58:742–748.[Abstract/Free Full Text]
  5. Geurts JJ, Bo L, Pouwels PJ, Castelijns JA, Polman CH, Barkhof F. Cortical lesions in multiple sclerosis: combined postmortem MR imaging and histopathology. AJNR Am J Neuroradiol 2005;26:572–577.[Abstract/Free Full Text]
  6. Cercignani M, Inglese M, Pagani E, Comi G, Filippi M. Mean diffusivity and fractional anisotropy histograms of patients with multiple sclerosis. AJNR Am J Neuroradiol 2001;22:952–958.[Abstract/Free Full Text]
  7. Rashid W, Hadjiprocopis A, Griffin CM, et al. Diffusion tensor imaging of early relapsing-remitting multiple sclerosis with histogram analysis using automated segmentation and brain volume correction. Mult Scler 2004;10:9–15.[Abstract/Free Full Text]
  8. Rovaris M, Bozzali M, Iannucci G, et al. Assessment of normal-appearing white and gray matter in patients with primary progressive multiple sclerosis: a diffusion-tensor magnetic resonance imaging study. Arch Neurol 2002;59:1406–1412.[Abstract/Free Full Text]
  9. Bozzali M, Cercignani M, Sormani MP, Comi G, Filippi M. Quantification of brain gray matter damage in different MS phenotypes by use of diffusion tensor MR imaging. AJNR Am J Neuroradiol 2002;23:985–988.[Abstract/Free Full Text]
  10. Vrenken H, Pouwels PJ, Geurts JJ, et al. Altered diffusion tensor in multiple sclerosis normal-appearing brain tissue: cortical diffusion changes seem related to clinical deterioration. J Magn Reson Imaging 2006;23:628–636.[CrossRef][Medline]
  11. Oreja-Guevara C, Rovaris M, Iannucci G, et al. Progressive gray matter damage in patients with relapsing-remitting multiple sclerosis: a longitudinal diffusion tensor magnetic resonance imaging study. Arch Neurol 2005;62:578–584.[Abstract/Free Full Text]
  12. Griffin CM, Chard DT, Ciccarelli O, et al. Diffusion tensor imaging in early relapsing-remitting multiple sclerosis. Mult Scler 2001;7:290–297.[Abstract/Free Full Text]
  13. Redpath TW, Smith FW. Technical note: use of a double inversion recovery pulse sequence to image selectively grey or white brain matter. Br J Radiol 1994;67:1258–1263.[Abstract/Free Full Text]
  14. Turetschek K, Wunderbaldinger P, Bankier AA, et al. Double inversion recovery imaging of the brain: initial experience and comparison with fluid attenuated inversion recovery imaging. Magn Reson Imaging 1998;16:127–135.[CrossRef][Medline]
  15. Bedell BJ, Narayana PA. Implementation and evaluation of a new pulse sequence for rapid acquisition of double inversion recovery images for simultaneous suppression of white matter and CSF. J Magn Reson Imaging 1998;8:544–547.[Medline]
  16. Geurts JJ, Pouwels PJ, Uitdehaag BM, Polman CH, Barkhof F, Castelijns JA. Intracortical lesions in multiple sclerosis: improved detection with 3D double inversion-recovery MR imaging. Radiology 2005;236:254–260.[Abstract/Free Full Text]
  17. Pouwels PJ, Kuijer JP, Mugler JP 3rd, Guttmann CR, Barkhof F. Human gray matter: feasibility of single-slab 3D double inversion-recovery high-spatial-resolution MR imaging. Radiology 2006;241:873–879.[Abstract/Free Full Text]
  18. Nelson F, Poonawalla AH, Hou P, Huang F, Wolinsky JS, Narayana P. Improved visualization of intracortical lesions in multiple sclerosis by phase-sensitive inversion recovery in combination with fast double inversion recovery MR imaging [abstr]. In: Twenty-second Congress of the European Committee for Treatment and Research in Multiple Sclerosis, Madrid, Spain. Basel, Switzerland: European Committee for Treatment and Research in Multiple Sclerosis, 2006; 639.
  19. Hou P, Hasan KM, Sitton CW, Wolinsky JS, Narayana PA. Phase-sensitive T1 inversion recovery imaging: a time-efficient interleaved technique for improved tissue contrast in neuroimaging. AJNR Am J Neuroradiol 2005;26:1432–1438.[Abstract/Free Full Text]
  20. Jezzard P, Barnett AS, Pierpaoli C. Characterization of and correction for eddy current artifacts in echo planar diffusion imaging. Magn Reson Med 1998;39:801–812.[Medline]
  21. Tao G, He R, Poonawalla AH, Narayana PA. Nonlinear image registration with voxel adaptive regularization to correct for EPI-induced geometric distortions [abstr]. In: Proceedings of the 15th International Society for Magnetic Resonance in Medicine, Berlin, Germany. Berkley, Calif: International Society for Magnetic Resonance in Medicine, 2007; 567.
  22. Madi S, Hasan KM, Narayana PA. Diffusion tensor imaging of in vivo and excised rat spinal cord at 7 T with an icosahedral encoding scheme. Magn Reson Med 2005;53:118–125.[CrossRef][Medline]
  23. Sajja BR, Datta S, He R, et al. Unified approach for multiple sclerosis lesion segmentation on brain MRI. Ann Biomed Eng 2006;34:142–151.[CrossRef][Medline]
  24. Datta S, Sajja BR, He R, Wolinsky JS, Gupta RK, Narayana PA. Segmentation and quantification of black holes in multiple sclerosis. Neuroimage 2006;29:467–474.[CrossRef][Medline]
  25. Hasan KM, Gupta RK, Santos RM, Wolinsky JS, Narayana PA. Diffusion tensor fractional anisotropy of the normal-appearing seven segments of the corpus callosum in healthy adults and relapsing-remitting multiple sclerosis patients. J Magn Reson Imaging 2005;21:735–743.[CrossRef][Medline]
  26. Peterson JW, Bo L, Mork S, Chang A, Trapp BD. Transected neurites, apoptotic neurons, and reduced inflammation in cortical multiple sclerosis lesions. Ann Neurol 2001;50:389–400.[CrossRef][Medline]
  27. McKinstry RC, Mathur A, Miller JH, et al. Radial organization of developing preterm human cerebral cortex revealed by non-invasive water diffusion anisotropy MRI. Cereb Cortex 2002;12:1237–1243.[Abstract/Free Full Text]
  28. Douaud G, Poupon C, Cointepas Y, et al. Diffusion tensor imaging (DTI) in Huntington's disease patients: analyses of fractional anisotropy (FA) maps and apparent diffusion coefficient (ADC) maps. In: Proceedings of the International Society for Magnetic Resonance in Medicine Workshop on Methods for Quantitative Diffusion MRI of Human Brain, Lake Louise, Canada, 2005.
  29. Bhagat YA, Beaulieu C. Diffusion anisotropy in subcortical white matter and cortical gray matter: changes with aging and the role of CSF-suppression. J Magn Reson Imaging 2004;20:216–227.[CrossRef][Medline]
  30. Batchelor PG, Atkinson D, Hill DL, Calamante F, Connelly A. Anisotropic noise propagation in diffusion tensor MRI sampling schemes. Magn Reson Med 2003;49:1143–1151.[CrossRef][Medline]
  31. Poonawalla AH, Zhou XJ. Analytical error propagation in diffusion anisotropy calculations. J Magn Reson Imaging 2004;19:489–498.[CrossRef][Medline]
  32. Falconer JC, Narayana PA. Cerebrospinal fluid-suppressed high-resolution diffusion imaging of human brain. Magn Reson Med 1997;37:119–123.[Medline]
  33. Woods RP, Grafton ST, Watson JD, Sicotte NL, Mazziotta JC. Automated image registration. II. Intersubject validation of linear and nonlinear models. J Comput Assist Tomogr 1998;22:153–165.



This article has been cited by other articles:


Home page
 BrainHome page
Y. He, A. Dagher, Z. Chen, A. Charil, A. Zijdenbos, K. Worsley, and A. Evans
Impaired small-world efficiency in structural cortical networks in multiple sclerosis associated with white matter lesion load
Brain, May 12, 2009; (2009) awp089v1.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2463070486v1
246/3/880    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Poonawalla, A. H.
Right arrow Articles by Narayana, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Poonawalla, A. H.
Right arrow Articles by Narayana, P. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE