| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 From the Department of Radiology (A.C.G., J.R.P., J.M.P.) and the Division of Hematology-Oncology, Department of Pediatrics (J.K.), Box 3808, Duke University Medical Center, Durham, NC 27710. A portion of this article was presented at the 1999 RSNA scientific assembly. Received May 3, 2000; revision requested June 29; revision received September 1; accepted September 14. Address correspondence to A.C.G. (e-mail: acguo@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: In eight patients with Krabbe disease and eight age-matched control subjects, anisotropy maps were generated with diffusion tensor data by using echo-planar imaging with diffusion gradient encoding in six directions. Anisotropy maps and T2-weighted images were visually inspected. Relative anisotropy (RA) and normalized T2-weighted signal intensity in white matter tracts and gray matter nuclei were quantitatively compared between patients and controls (paired Student t test).
RESULTS: Loss of diffusion anisotropy appeared on anisotropy maps as areas of decreased hyperintensity in patients with Krabbe disease. Differences in RA between Krabbe disease patients and control subjects were significant in eight of nine white matter structures studied (P = .001–.01) and in basal ganglia (P = .04). T2-weighted signal intensity was also significantly different in the same white matter structures (P = .006–.049) but not in basal ganglia. In the three patients imaged after stem cell transplantation, mean RA was between the RAs of untreated patients and control subjects.
CONCLUSION: Diffusion tensor–derived anisotropy maps (a) provide a quantitative measure of abnormal white matter in patients with Krabbe disease, (b) are more sensitive than T2-weighted images for detecting white matter abnormality, and (c) may be a marker of treatment response.
Index terms: Brain, diffusion, 10.8732 • Brain, diseases, 10.8732 • Brain, metabolism, 10.8732 • Brain, MR, 10.121411, 10.121416, 10.12149 • Brain, white matter, 10.8732
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In the past, investigators have attempted to characterize anisotropic diffusion with conventional (trace-weighted) diffusion-weighted magnetic resonance (MR) imaging by measuring apparent diffusion coefficients in two or three perpendicular directions along the magnetic gradients (1–4). This method is inherently biased toward diffusion along the gradients employed and generally underestimates the degree of diffusion anisotropy (2). In contrast, diffusion tensor MR imaging, which is a more recently developed technique, provides a more complete description of water self-diffusion than trace-weighted diffusion imaging by providing information with regard to not only the overall magnitude of diffusion but also the degree of diffusion anisotropy. Therefore, diffusion tensor MR imaging offers a potentially more sensitive means for the detection of white matter disease.
Krabbe disease, also known as globoid cell leukodystrophy, is an autosomal recessive disorder that results from deficiency of the lysosomal enzyme, galactocerebroside ß-galactosidase (also known as ß-galactocerebrosidase and galactosylceramide ß-galactosidase) (5–8). Deficiency of this enzyme blocks the degradation of ß-galactocerebroside, a major component of the myelin sheath (6). In addition, catabolic derivatives of ß-galactocerebroside such as psychosine, which is toxic to oligodendrocytes, can accumulate (6,8). The diagnosis is generally based on a combination of clinical, radiologic, and biochemical findings and is confirmed with a biochemical assay of leukocyte lysosomal ß-galactosidase levels (5–9).
In general, Krabbe disease occurs in infants and young children and predominantly affects the white matter tracts of the central nervous system (6,8,10). At MR imaging, the pyramidal tract, periventricular white matter, cerebellar white matter, and deep gray matter are most often affected in the early-onset form, and atrophy becomes prominent later in the course of the disease (9). In the late-onset form, the parietooccipital white matter and the posterior corpus callosum, in addition to the pyramidal tracts, are more frequently involved, and cerebellar white matter and deep gray matter are relatively spared (9). In addition, patients with the late-onset form usually develop little atrophy (9).
Until recently, essentially no treatment existed for Krabbe disease. However, in 1998, investigators reported encouraging results from a study of five children treated with hematopoietic stem cell transplantation, with both clinical and radiologic manifestations of the disease either reversed or retarded (11). These results made early detection and monitoring of the disease course with imaging increasingly important. Given the potential sensitivity of diffusion tensor MR imaging in the assessment of white matter disease, we set out to compare diffusion tensor MR imaging with conventional T2-weighted MR imaging for the evaluation of white matter changes in patients with Krabbe disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Control Subjects
The data from eight children (five boys, three girls; age range, 5 weeks to 3 years) who had undergone MR examinations performed within the same period were chosen as control data. The control subjects were chosen on the basis of (a) closest age-match with patients with Krabbe disease and (b) normal findings with the routine MR sequences (results of MR examination reported as normal). Because only three of the five clinical MR imagers at our institution had the capability to be used for diffusion tensor MR imaging during this period, only children who had undergone diffusion tensor MR imaging, in addition to having normal findings at routine MR examination, were considered. Indications for imaging of control subjects included seizures (n = 3), autism (n = 1), facial hemangioma (n = 1), scalp lesion (n = 1), torticollis (n = 1), and subcutaneous lipoma (n = 1). Children who were referred because they were suspected of having white matter disease were specifically excluded.
MR Data Acquisition and Analysis
All subjects (patients and controls) were imaged on 1.5-T clinical MR imagers (Signa; GE Medical Systems, Milwaukee, Wis) with a standard head coil. Both diffusion tensor MR imaging and routine MR imaging were performed in all subjects. The diffusion tensor protocol consisted of one-shot spin-echo echo-planar imaging with a repetition time (TR) of 12,000 msec, inversion time (TI) of 2,200 msec, minimum echo time (TE), and one signal acquired, with diffusion gradient encoding in six directions by using b = 1,000 sec/mm2, as well as by using no diffusion gradient (b = 0). Therefore, a total of seven diffusion-weighted images were obtained for each image section. Section thickness was 4 mm, and section gap was 2 mm. The field of view was 40 cm x 20 cm with a 128 x 64 matrix. Imaging time for the diffusion tensor sequence was approximately 2 minutes. Routine MR sequences that were used were tailored according to the clinical manifestation but in all cases included a spin-echo T2-weighted sequence (2,800/100 [TR msec/TE msec]; field of view, 22 cm2; matrix size, 256 x 192 [frequency direction x phase direction]; section thickness, 4 mm; section gap, 2 mm; two signals acquired).
The raw diffusion image data were transferred to an independent workstation (Advantage Windows; GE Medical Systems) and processed by using a commercial software program (FUNCTOOL; GE Medical Systems) as well as proprietary software. The six independent elements of the diffusion tensor D (ie, Dxx, Dyy, Dzz, Dxy, Dxz, and Dyz) were statistically calculated for each voxel on the basis of the method described by Basser et al (12–14) and on the basis of the following equation: ln[A(b)/A(b=0)] = 
ijbijDij, where bij is the component of the ith row and jth column of the diffusion gradient matrix b; A(b) is the resultant echo intensity for a gradient sequence with directions and magnitudes of the diffusion-sensitizing gradients described by the b matrix; A(b=0) is the echo intensity when b is the zero matrix (no diffusion gradient); and Dij is the corresponding component of the diffusion tensor matrix D. Once the elements of the diffusion tensor were determined, its eigenvalues were obtained by using diagonalization of the tensor matrix. The trace (which can be thought of as the mean diffusivity) was then calculated from the eigenvalues.
For the index of anisotropy, we chose to use the variance of the eigenvalues because it is a function solely of the diffusion tensor and is independent of the ordering scheme of eigenvalues. Therefore, it is rotationally invariant and less susceptible to noise than some of the other proposed indices, which depend on ordering of eigenvalues (13). In addition, the variance of eigenvalues is a good representation of the ratio of the anisotropic component to the isotropic component of diffusion, that is, relative anisotropy (RA), and provides relatively good contrast between gray matter and white matter (13,15). The equation used for RA (the variance) is as follows: V = 1/31/2 [(E1-d)2 + (E2-d)2 + (E3-d)2]1/2/d, where Ei = eigenvalues, and d = (E1+E2+E3)/3. These calculations were performed for each voxel and displayed as an anisotropy map.
Two neuroradiologists (A.C.G., J.M.P.) who were not blinded to the patients reviewed the relative anisotropy maps and T2-weighted MR images for conspicuity of abnormalities in major white matter regions of the brain. A qualitative comparison of patients and controls was made, and the impression was recorded for each patient-control pair. When there was disagreement in the impression, a consensus was reached. For the treated patients, T2-weighted images obtained before and after treatment were reviewed and compared. However, pretreatment imaging in these patients was performed prior to the availability of diffusion tensor MR imaging at our institution. Therefore, no pretreatment anisotropy maps were obtained for comparison. The purpose of visual inspection was only to provide a qualitative estimate of lesion conspicuity and to guide placement of regions of interest (ROIs). Therefore, statistical analysis of visual impressions was not performed.
The ROIs were manually drawn by a single neuroradiologist (A.C.G.) who was not blinded to the patient’s identity or to the diagnosis. The ROIs were drawn on both anisotropy maps and T2-weighted MR images in each of the following 11 brain structures known to be affected in Krabbe disease: middle cerebellar peduncle, cerebral peduncle, thalamus, basal ganglia, internal capsule, genu and splenium of corpus callosum, frontal and parietooccipital white matter, corona radiata, and centrum semiovale (Fig 1) (5–10). In addition to the white matter structures, deep gray matter nuclei of the basal ganglia and thalamus were included because they are known to be involved early in the course of Krabbe disease (6,8). All ROIs were of uniform size (100 mm2 ± 7) except those in the cerebral peduncles, where smaller ROIs (40 mm2 ± 3) were needed to fit within smaller structures. The ROIs varied in shape to fit the structure of interest and were placed bilaterally, one on each side, for each structure.
|
|
For each of the 11 brain structures measured, a comparison was made between anisotropy values of patients with Krabbe disease and their age-matched controls by applying a paired Student t test. The same comparison was also made between T2-weighted signal intensities of patients and controls. A P value of less than .05 was considered to be significant. The P values for RA were compared with those for T2-weighted signal intensity. The number of patients included for each analyzed structure varied from six to eight, and the inclusion criteria were based on the age of the patient and whether the structure of interest was expected to be myelinated at that age. Specifically, the 3-week-old and 3-month-old patients were excluded from analysis of the genu of corpus callosum, frontal white matter, occipital white matter, and centrum semiovale because these structures are thought to not be well myelinated at 3 months of age (16). In addition, the 3-week-old patient was also excluded from analysis of the middle cerebellar peduncles, cerebral peduncles, and the splenium of the corpus callosum for the same reason.
An additional comparison was made between the 3-week-old patient, the youngest in the study, and her control by treating all ROIs in the patient as one population and all ROIs in the control as another population. This method is in contrast to treating our group of Krabbe patients as a population, as was stated in the previous paragraph. A paired Student t test was performed by pairing each of the 11 ROIs in this 3-week-old patient with the same ROIs in her control. A P value of less than .05 was considered to be significant. This additional comparison was performed because we expected lesion detection to be difficult in the youngest patients because of their minimal degree of white matter myelination. However, the comparison was done with the understanding that the results are applicable only to this specific patient-control pair and cannot be generalized to our population of patients with Krabbe disease.
Because the number of treated patients was too small to statistically analyze separately, the data from patients who had been treated with hematopoietic stem cell transplantation (n = 3) were analyzed in conjunction with those who had not been treated (n = 5). This was done with the understanding that treated patients would be expected to have a lesser degree of disease involvement on the basis of both clinical findings and the findings at conventional MR imaging. Therefore, the inclusion of treated patients would favor the null hypothesis (ie, that there is no anisotropy difference detectable with diffusion tensor MR imaging), and any difference detected between patients with Krabbe disease and controls would have been more evident in a population consisting exclusively of untreated patients. However, mean anisotropy values of the three treated patients were plotted against data from three untreated patients and three controls of similar ages to assess for trends.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
1 year old) and were less evident in the younger patients. In the youngest patient (age, 3 weeks), differences were not detectable visually on either T2-weighted MR images or anisotropy maps.
|
|
|
|
Quantitative Assessment: All Patients
Significant (P < .05) differences in RA were observed between patients with Krabbe disease (both before and after treatment) and age-matched controls in nine of the 11 anatomic structures studied, including the basal ganglia, middle cerebellar peduncles, posterior limbs of internal capsule, genu of corpus callosum, splenium of corpus callosum, frontal white matter, occipital white matter, corona radiata, and centrum semiovale. No significant difference was observed in the cerebral peduncles (P = .25) or thalami (P = .14).
In comparison, significant differences in T2-weighted signal intensity were observed between the same patients and their controls in eight of the 11 anatomic structures studied, including the middle cerebellar peduncles, posterior limbs of internal capsule, genu of corpus callosum, splenium of corpus callosum, frontal white matter, occipital white matter, corona radiata, and centrum semiovale. No significant difference was observed in the cerebral peduncles (P = .39), basal ganglia (P = .10), and thalami (P = .64). The P values for T2-weighted signal intensity were, in general, in the range of 10-2 versus in the range of 10-3 for RA, or approximately one order of magnitude greater (Tables 1, 2). There was an overall difference in RA between the youngest patient and her control (P = .03, two-tailed paired t test comparing all ROIs in a solitary patient-control pair), despite the lack of differences at visual inspection.
|
|
Quantitative Assessment: Treated Patients
Because of the small number of treated patients (n = 3), the mean of their data was calculated (for each anatomic structure) and plotted against mean data from an equal number of untreated patients and an equal number of controls on a linear graph (Fig 4). Despite the nearly normal appearance of white matter on both T2-weighted images and anisotropy maps in two of the three patients, the mean values for RA in treated patients were consistently lower than those in controls and higher than those in untreated patients in all assessed white matter structures (even when values from the poorly responding third treated patient were included with the other treated patients to calculate the mean). In most white matter structures, the mean RA of treated patients was approximately halfway between that of untreated patients and controls.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The ranges of RA measured in this study for both normal gray and white matter were similar to the ranges measured in other studies with similar indices for anisotropy (Table 1) (15,17,18). The results in diseased white matter were also consistent with findings of previous studies that applied diffusion tensor MR imaging to other white matter diseases, such as multiple sclerosis and amyotrophic lateral sclerosis, in that a marked decrease in anisotropy was observed (19,20). The only white matter tract in which a significant difference was not detected was the cerebral peduncle, where RA was relatively low in both patients with Krabbe disease and controls. This fact is not surprising, given that the cerebral peduncle is a relatively small structure surrounded by cerebrospinal fluid on three sides, which increases the likelihood of volume averaging. The density of fibers in the cerebral peduncle is less than 20% of the density in other major white matter tracts such as the corpus callosum and anterior commissure (21), a fact that may also have contributed to the relatively low anisotropy measured in this structure.
We compared the sensitivity of diffusion anisotropy maps with T2-weighted MR images for the detection of abnormal white matter because T2-weighted images have been the standard images for evaluation of white matter disease. We found that P values for RA were consistently smaller than those for T2-weighted signal intensity across all anatomic structures measured, often by an order of magnitude (Tables 1, 2). The lower P values associated with RA suggest that anisotropy maps may be more sensitive than T2-weighted images in the detection of white matter disease processes. The increased sensitivity of anisotropy maps may be attributable to the fact that changes in anisotropy are a better measurement of the primary disease process (ie, dysmyelination and demyelination) (1,2).
The difference in anisotropy between patients with Krabbe disease and controls was not only evident with quantitative analysis but was also evident with visual inspection in a majority of our patients (Figs 2, 3). However, in certain subgroups of patients where differences were subtle, quantification provided a more sensitive means for detection of disease. For example, in two of three patients who underwent hematopoietic stem cell transplantation, both T2-weighted images and anisotropy maps were only mildly abnormal at visual inspection, but their RAs were markedly lower than those in controls. Usually, the RAs for treated patients were midway between untreated patients and controls (Fig 4), which suggested that visual inspection alone may be inadequate for the detection of pathologic alterations in white matter.
An additional observation in treated patients is that all three patients have shown either improvement or a substantial delay in disease progression on conventional MR images, as well as clinical improvement, compared with other patients who had similar age of onset and clinical features. Therefore, findings on anisotropy maps may correlate with the effect of treatment. However, because these patients did not undergo diffusion tensor MR imaging prior to transplantation, the possibility of selection bias cannot be excluded as an explanation for the relatively preserved anisotropy compared with untreated patients.
The sensitivity of anisotropy maps compared with conventional MR imaging was also illustrated by the results from the youngest patient enrolled in the study. No discernible difference was seen on T2-weighted images between this patient and her control. However, when RA was compared across all white matter regions, a significant difference (P = .03) was found. The quantitative features of diffusion tensor MR imaging may prove to be a valuable means of evaluating white matter disease processes in situations in which visual assessment might prove insensitive for detection of subtle changes.
The RA in the basal ganglia was also significantly lower in patients with Krabbe disease than in controls, although T2-weighted signal intensity did not show a similar significant difference in the basal ganglia (P = .10). On the basis of our experience and the experience of other investigators, gray matter anisotropy is 20%–60% of that in white matter (15,17,18). This may be due to the less ordered nature of the gray matter microenvironment (16,18). It is known that both the basal ganglia and thalamus are affected relatively early in the course of Krabbe disease, but the underlying pathophysiologic function is unclear (8). Some investigators have suggested as possible causes the transient calcification and alterations of lipid, water, and protein ratios as a result of myelin breakdown (8,11), and these factors may also alter diffusion anisotropy in deep gray matter nuclei. Because abnormality of the basal ganglia is an early finding in patients with Krabbe disease and is often difficult to detect with routine MR sequences (8), diffusion tensor MR imaging may provide an advantage for early disease detection in this respect. Nonetheless, it is possible that given the relatively small patient sample and the relatively high P value, finding a significant anisotropy difference in the basal ganglia was somewhat fortuitous.
A few limitations to this study should be noted. Although our study is one of the largest reported series of patients with Krabbe disease, the patient population is nonetheless small because of the rarity of this disease. The small sample size limits our ability to generalize our results to the general population of patients with Krabbe disease. The small sample size also prevents us from analyzing the subgroups of treated patients, untreated patients, and different age groups separately. We intend to extend the study to include a sufficient sample size to address these issues. A second limitation is the lack of longitudinal data. We were unable to determine whether anisotropy values of treated patients were improved compared with baseline. Therefore, it is difficult to establish the effect of treatment. We also did not compare diffusion tensor MR imaging with other techniques such as fluid-attenuated inversion-recovery and trace-weighted diffusion-weighted sequences for sensitivity of lesion detection. A final limitation lies in the relatively low signal-to-noise and contrast-to-noise ratios achieved by using a single image acquisition. Using multiple acquisitions in the future should improve measurement of anisotropy.
In conclusion, anisotropy maps and indices derived from diffusion tensor MR imaging provide a sensitive and quantitative means for the detection of abnormal white matter in patients with Krabbe disease. Our findings suggest that these maps and indices are more sensitive than the findings from T2-weighted MR imaging, which has been the standard technique for evaluation of white matter disease. Although diffusion tensor MR imaging was not necessary for making the diagnosis of Krabbe disease in our patients, we were able to detect subtle differences between patients and controls not evident on T2-weighted images. In addition, our preliminary data suggest that diffusion tensor MR imaging may better define the effects of treatment in patients who have undergone hematopoietic stem cell transplantation. As new treatment modalities become available for previously untreatable diseases, the ability to detect subtle early findings, to assess disease progression, and to follow the effects of treatment has become increasingly important. Because diffusion tensor MR imaging can be used to assess and quantify anisotropy, a physiologic parameter that closely reflects structural organization in white matter, this technique holds promise for the evaluation of Krabbe disease as well as other leukodystrophies.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, A.C.G., J.M.P.; study concepts, all authors; study design, A.C.G., J.R.P., J.M.P.; definition of intellectual content, all authors; literature research, A.C.G.; clinical studies, A.C.G., J.M.P.; data acquisition, A.C.G.; data analysis, A.C.G., J.R.P., J.M.P.; statistical analysis, A.C.G., J.R.P.; manuscript preparation, A.C.G., J.M.P.; manuscript editing, A.C.G., J.R.P., J.M.P.; manuscript review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. M. Provenzale, L. Liang, D. DeLong, and L. E. White Diffusion Tensor Imaging Assessment of Brain White Matter Maturation During the First Postnatal Year Am. J. Roentgenol., August 1, 2007; 189(2): 476 - 486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. van der Voorn, P. J. W. Pouwels, A. A. M. Hart, J. Serrarens, M. A. A. P. Willemsen, H. P. H. Kremer, F. Barkhof, and M. S. van der Knaap Childhood White Matter Disorders: Quantitative MR Imaging and Spectroscopy. Radiology, November 1, 2006; 241(2): 510 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Provenzale, S. Mukundan, and D. P. Barboriak Diffusion-weighted and Perfusion MR Imaging for Brain Tumor Characterization and Assessment of Treatment Response. Radiology, June 1, 2006; 239(3): 632 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Hasan and J. Provenzale Diffusion Tensor Eigenvalues or Both Mean Diffusivity and Fractional Anisotropy Are Required in Quantitative Clinical Diffusion Tensor MR Reports: Fractional Anisotropy Alone Is Not Sufficient. Radiology, May 1, 2006; 239(2): 611 - 613. [Full Text] [PDF]
|
|
|
|
|
|
P. McGraw, L. Liang, M. Escolar, S. Mukundan, J. Kurtzberg, and J. M. Provenzale Krabbe Disease Treated with Hematopoietic Stem Cell Transplantation: Serial Assessment of Anisotropy Measurements--Initial Experience Radiology, July 1, 2005; 236(1): 221 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shiraishi, Y. Hasegawa, S. Okada, K. Kimura, T. Sawada, H. Mizusawa, and K. Minematsu Highly Diffusion-Sensitized Tensor Imaging of Unilateral Cerebral Arterial Occlusive Disease AJNR Am. J. Neuroradiol., June 1, 2005; 26(6): 1498 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-F. Tu, C.-Y. Chen, C.-C. Huang, and C.-S. Lee Vacuolating Megalencephalic Leukoencephalopathy with Mild Clinical Course Validated by Diffusion Tensor Imaging and MR Spectroscopy AJNR Am. J. Neuroradiol., June 1, 2004; 25(6): 1041 - 1045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chen, S.-K. Song, W. Liu, M. McLean, J. S. Allen, J. Tan, S. A. Wickline, and X. Yu Remodeling of cardiac fiber structure after infarction in rats quantified with diffusion tensor MRI Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H946 - H954. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.-L. Khong, D. L. W. Kwong, G. C. F. Chan, J. S. T. Sham, F.-L. Chan, and G.-C. Ooi Diffusion-Tensor Imaging for the Detection and Quantification of Treatment-Induced White Matter Injury in Children with Medulloblastoma: A Pilot Study AJNR Am. J. Neuroradiol., April 1, 2003; 24(4): 734 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Guo, J. R. MacFall, and J. M. Provenzale Multiple Sclerosis: Diffusion Tensor MR Imaging for Evaluation of Normal-appearing White Matter Radiology, March 1, 2002; 222(3): 729 - 736. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Melhem, S. Mori, G. Mukundan, M. A. Kraut, M. G. Pomper, and P. C. M. van Zijl Diffusion Tensor MR Imaging of the Brain and White Matter Tractography Am. J. Roentgenol., January 1, 2002; 178(1): 3 - 16. [Full Text] [PDF]
|
|
|
|
|
|
A. C. Guo, V. L. Jewells, and J. M. Provenzale Analysis of Normal-Appearing White Matter in Multiple Sclerosis: Comparison of Diffusion Tensor MR Imaging and Magnetization Transfer Imaging AJNR Am. J. Neuroradiol., November 1, 2001; 22(10): 1893 - 1900. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY< |
