| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Neuroradiology |
1 From Depts of Neurosurgery (F.Y., K.K., K.A., K. Sugiyama, A.T., R.H., S.H., Y.K., K.Y., T.S., M.A.T.) and Radiology (J.T.), Graduate School of Biomedical Sciences, Hiroshima Univ, 1–2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan; Dept of Environmetrics and Biometrics, Research Inst for Radiation Biology and Medicine, Hiroshima Univ, Japan (K. Satoh, M.O.); and Dept of Neurosurgery, National Hosp, Kure Med Ctr, Kure, Japan (H.Y., Y.I.). Received Aug 21, 2003; revision requested Oct 31; final revision received Aug 19, 2004; accepted Sep 8. Address correspondence to K.K. (e-mail: kuka422@hiroshima-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Institutional review board approval or informed patient consent was not required. MR images were reviewed retrospectively in 275 patients with brain tumors: 147 males and 128 females 1–81 years old, treated between September 1997 and July 2003. Regions of interest were placed manually in tumor regions on MR images, and ADC was calculated with a five-point regression method at b values of 0, 250, 500, 750, and 1000 sec/mm2. ADC values were average values in tumor. All brain tumor subgroups were analyzed. Logistic discriminant analysis was performed by using ADC, age, and patient sex as independent variables to discriminate among tumor groups.
RESULTS: A significant negative correlation existed between ADC and astrocytic tumors of World Health Organization grades 2–4 (grade 2 vs grades 3 and 4, accuracy of 91.3% [P < .01]; grade 3 vs 4, accuracy of 82.4% [P < .01]). ADC of dysembryoplastic neuroepithelial tumors (DNTs) was higher than that of astrocytic grade 2 tumors (accuracy, 100%) and other glioneuronal tumors. ADC of malignant lymphomas was lower than that of glioblastomas and metastatic tumors (accuracy, 83.6%; P < .01). ADC of primitive neuroectodermal tumors (PNETs) was lower than that of ependymomas (accuracy, 100%). ADC of meningiomas was lower than that of schwannomas (accuracy, 92.4%; P < .01). ADC of craniopharyngiomas was higher than that of pituitary adenomas (accuracy, 85.2%; P < .05). ADC of epidermoid tumors was lower than that of chordomas (accuracy, 100%). In meningiomas, ADC was not indicative of malignancy grade or histologic subtype.
CONCLUSION: ADC is useful for differentiation of some human brain tumors, particularly DNT, malignant lymphomas versus glioblastomas and metastatic tumors, and ependymomas versus PNETs.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MR Imaging and Image Processing
All MR studies were performed with a 1.5-T superconducting system (Signa Horizon; GE Medical Systems, Milwaukee, Wis) and a circularly polarized head coil. All patients underwent MR imaging, which included, at the minimum, unenhanced and contrast material–enhanced transverse T1-weighted images, unenhanced transverse T2-weighted images, unenhanced transverse fluid-attenuated inversion-recovery (FLAIR) images, and unenhanced transverse DW images. We (J.T. and F.Y., with 15 and 10 years of experience with brain MR imaging, respectively) performed these imaging studies by using the same section orientations for all examinations; selections were made by means of consensus.
The transverse T1-weighted spin-echo MR sequence was performed with the following parameters: repetition time msec/echo time msec, 400/8; field of view, 22 x 16 cm; matrix size, 256 (frequency) x 192 (phase); section thickness, 5 mm; section gap, 2.5 mm; and two signals acquired. The contrast-enhanced T1-weighted sequences were performed after the administration of a gadolinium compound (0.1 mmol gadodiamide per kilogram of body weight). The transverse fast spin-echo T2-weighted sequence was performed with the following parameters: 3500/100; field of view, 22 x 16 cm; matrix size, 256 x 192; echo train length, 12; section thickness, 5 mm; section gap, 2.5 mm; and two signals acquired. Transverse FLAIR images were obtained by using fast and interleaved multisection sequences with the following parameters: 10 000/150; inversion time, 2200 msec; field of view, 22 x 22 cm; matrix size, 256 x 192; echo train length, 16; section thickness, 5 mm; section gap, 2.5 mm; and one signal acquired.
Transverse DW imaging was performed by using a single-shot T2-weighted echo-planar spin-echo sequence with the following parameters: 1600/107; diffusion gradient encoding in three (x, y, z) orthogonal directions; b values of 250, 500, 750, and 1000 sec/mm2; field of view, 24 x 24 cm; matrix size, 128 x 128; section thickness, 7.5 mm; section gap, 0 mm; and one signal acquired. At each b value, x, y, and z single-direction DW images and a baseline image (b = 0 sec/mm2) were acquired; combined ([x + y + z]/3) DW imaging was calculated and performed automatically by the MR instrument. We obtained 10 sections with 50 images at each b value in 13 seconds (10 images of combined [{x + y + z}/3] DW imaging, 10 images of the baseline image, and 10 images each of the x-, y-, and z-direction DW images). Therefore, each DW imaging study yielded a total of 200 images.
All DW imaging data were transferred to a computer workstation (SUN Sparc 20; Sun Microsystems, Mountain View, Calif) for determination of the signal intensity and ADC. MR Vision (version 1.5.5; L.A. Systems, Oyama, Japan) was the software program used to generate ADC maps and quantify ADCs of brain tumors. To avoid the contrast from diffusion anisotropy, we selected only combined ([x + y + z]/3) DW images for the production of ADC maps by calculating the signal intensity of the DW images at five b values (0, 250, 500, 750, 1000) on a pixel-by-pixel basis. We used b = 0 images; these are obtained at the time the b = 250 images are acquired. Each image used for the creation of the ADC maps was obtained with one signal acquired.
The ADC was measured by manually placing regions of interest in tumor regions on the ADC map. In patients with contrast-enhanced tumors, regions of interest were placed at the site of enhanced lesions on contrast-enhanced T1-weighted MR images. In patients with weakly enhancing or nonenhancing tumors, regions of interest were chosen after identifying the tumor area as an area of hyperintensity on FLAIR images. Cystic components were differentiated as both areas of hyperintensity on T2-weighted MR images and areas of hypointensity on FLAIR MR images. Necrotic components were differentiated on contrast-enhanced T1-weighted images as the interior of enhanced lesions. Hemorrhagic lesions were differentiated on unenhanced T1-weighted MR images as areas of hyperintensity. We (F.Y., J.T., and Y.K., with 10, 15, and 9 years of brain MR imaging experience, respectively) compared the ADC maps and other MR images carefully and placed the regions of interest only in the solid tumor components by means of consensus. We excluded cystic, necrotic, and hemorrhagic tumor areas. We chose regions of interest as central as possible within the tumor area at random and averaged the ADC of each tumor.
Reference Standard
The ADC of individual tumors was determined preoperatively (F.Y., J.T., and Y.K.), and each tumor was graded according to the WHO system by two neuropathologists (K. Sugiyama and S.H., with 18 and 12 years of experience, respectively) in consensus, who were blinded to the ADC value and imaging information.
Statistical Analysis
Statistical analysis was performed with a commercially available software package (Statview, version 5.0; SAS Institute, Cary, NC). For each brain tumor classified according to the WHO system in our series, we calculated the mean and standard deviation of the ADC and patient age. Logistic discriminant (regression) analysis was performed by using ADC, age, and patient sex as independent variables X1, X2, and X3, respectively, to discriminate among the tumor groups. With two groups, the method assumes that the probabilities P0 and P1 of group membership follow the logistic model: P0 = exp(ß0 + ß1X1 + ß2X2 + ß3X3)/(1 + exp[ß0 + ß1X1 + ß2X2 + ß3X3]) and P1 = 1 – P0 = 1/(1 + exp[ß0 + ß1X1 + ß2X2 + ß3X3]), where ß0 is the constant, ß1 is the coefficient of X1 (which is the ADC [x 1000 m2/sec]), ß2 is the coefficient of X2 (which is patient age [in years]), and ß3 is the coefficient of X3 (which is patient sex [male = 0, female = 1]).
Note that in cases where the misclassification rate is zero, maximum log likelihood estimators and the significance of the estimated coefficients are not available (17). The histologic results for two compared tumor groups were defined as P0 and P1, respectively. The discriminant rate and diagnostic accuracy were calculated.
We compared each tumor group (P0 vs P1) as follows: low-grade (grades 1 and 2) versus high-grade (grades 3 and 4) glioneuronal tumors, grade 1 versus grade 2 glioneuronal tumors, grade 3 versus grade 4 glioneuronal tumors, grade 2 versus grades 3 and 4 astrocytic tumors, grade 3 versus grade 4 astrocytic tumors, grade 2 versus grade 3 oligodendrogliomas, grade 2 versus grade 3 ependymal tumors, grade 2 versus grade 3 astrocytic and oligodendroglial tumors, grade 2 astrocytic tumors versus grade 2 oligodendroglial tumors, grade 3 astrocytic tumors versus grade 3 oligodendroglial tumors, grade 2 versus grade 3 astrocytic and ependymal tumors, grade 2 astrocytic tumors versus ependymal tumors, grade 3 astrocytic tumors versus ependymal tumors, grade 2 astrocytic tumors versus DNTs, grade 1 glioneuronal tumors (except DNTs) versus DNTs, ependymal tumors versus primitive neuroectodermal tumors (PNETs; medulloblastomas, PNETs, and pineoblastomas), malignant lymphomas versus glioblastomas and metastatic tumors, glioblastomas versus metastatic tumors, grade 1 benign-subtype meningiomas versus grades 2 and 3 malignant subtype meningiomas, meningiomas versus schwannomas, pituitary adenomas versus craniopharyngiomas, pituitary adenomas versus meningiomas, pituitary adenomas versus germ cell tumors, craniopharyngiomas versus meningiomas, germ cell tumors versus craniopharyngiomas, meningiomas versus germ cell tumors, PNETs versus germ cell tumors, PNETs versus glioblastomas, PNETs versus meningiomas, PNETs versus malignant lymphomas, glioblastomas versus germ cell tumors, glioblastomas versus meningiomas, and epidermoid tumors versus chordomas.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
Our results show that we could discriminate without misclassification between grade 2 oligodendroglial tumors and grade 3 oligodendrogliomas, between grade 2 and 3 ependymal tumors, between grade 2 astrocytic tumors and ependymal tumors, and between grade 3 astrocytic tumors and ependymal tumors, although the number of these cases in our series was not large. On the other hand, we could not discriminate between astrocytic tumors and oligodendroglial tumors.
DNTs.—The ADC of DNTs is higher than that of diffuse astrocytomas, and the ADC of DNTs is higher than that of other grade 1 glioneuronal tumors. In addition, the ADC of DNTs is higher than that of any other neuroepithelial tumor, and there is no overlapping of ADC values.
Ependymoma versus PNET, Ventricle Tumors
We found that the ADC of ependymomas is higher than that of PNETs, and there is no overlapping. The ADC of ependymomas is consistently higher than 1.00 x 10–3 mm2/sec. The ADC of PNETs is consistently lower than 1.00 x 10–3 mm2/sec. The ADC of central neurocytomas is as low as that of glioblastomas and lower than that of subependymomas.
Glioblastomas, Metastatic Tumors, and Malignant Lymphomas
Malignant lymphomas manifest lower ADCs than those in glioblastomas and metastatic tumors (P < .01). In our study, however, glioblastomas and metastatic tumors could not be discriminated.
Meningiomas and Schwannomas
The ADC does not identify the histologically benign subtypes of meningioma. There was no statistical difference between the benign (WHO grade 1) and malignant (WHO grades 2 and 3) subtypes. In our series, there was only one microcystic meningioma. Its very high ADC may be characteristic of these tumors.
The ADC of schwannomas is significantly higher than that of meningiomas (P < .01).
Tumors Developing in Parasellar Lesions
Pituitary adenomas, craniopharyngiomas, germ cell tumors, and parasellar meningiomas that frequently develop in parasellar regions are sometimes difficult to differentiate from each other. We included only large adenomas with supratentorial extension because the microadenomas and small adenomas in our series were not sufficiently large to assign a region of interest. The ADC of pituitary adenomas is not significantly different from that of menigiomas and germ cell tumors. The ADC of craniopharyngiomas is significantly higher than that of pituitary adenomas (P < .05) and meningiomas (P < .05). We could discriminate without misclassification between craniopharyngiomas and germ cell tumors, and while the ADC distribution of craniopharyngiomas was higher than that of germ cell tumors, our series contained only a few cases. The ADC of germ cell tumors is not significantly different from that of meningiomas.
Tumors Developing Pineal Lesions and Other Analyses
The accuracy of logistic discriminant analysis is more than 90% between PNET and germ cell tumors, between PNET and glioblastomas, and between PNET and meningiomas. While the ADC distribution of PNET is lower than in those other tumors, a statistical difference was found only between PNET and glioblastomas (P < .05). We can discriminate without misclassification between PNET and malignant lymphomas; however, there is no difference in the ADC distribution between PNET and malignant lymphomas.
The ADC of epidermoid tumors is lower than that of chordomas; however, only a small number of cases were available for analysis.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge, the ADCs of ependymomas, medulloblastomas, and PNETs have not been compared to date. We found that the ADC of ependymomas was higher than that of PNETs, and there was no overlapping. Since the ADC of ependymomas was consistently higher than 1 x 10–3 mm2/sec and that of PNET was consistently lower than 1 x 10–3 mm2/sec, we suggest that preoperative determination of the ADC of fourth-ventricle tumors makes possible the differential diagnosis between ependymomas and medulloblastomas.
In our search of the literature, we were unable to find reports on the ADC of subependymomas and central neurocytomas. Since we found the ADC of central neurocytomas to be much lower than that of subependymomas, we posit that preoperative evaluation of the ADC of tumors at the lateral ventricle may provide useful information for differentiating between central neurocytomas and subependymomas. However, a study of patient populations larger than ours is necessary to determine the ADCs of rare tumors, such as central neurocytomas, pleomorphic xantoastrocytomas, gangliogliomas, subependymomas, and other rare neuroepithelial tumors.
DNTs
DNTs may develop in any part of the supratentorial cortex and may arise in the area of the caudate nucleus, cerebellum, third ventricles and pons, and septum pellucidum. The preoperative diagnosis of DNT tends to be difficult because these tumors exhibit nonspecific features on conventional MR and computed tomographic images. The ADC of DNT was higher than that of other WHO grade 1 and grade 2 gliomas. In addition, the ADC of DNTs was higher than that of any other glioneuronal tumors, and there was no overlapping of ADC values. The high ADC of DNTs may be attributable to the presence of large extracellular spaces and their cellularity, which is much lower than that of other human brain tumors. We suggest that preoperative ADC determination may facilitate a differential diagnosis of DNTs. The preoperative identification of DNTs and the distinction of DNTs from other gliomas have important treatment implications. During long-term follow-up, patients who had undergone complete or incomplete surgical removal of DNTs did not manifest clinical or radiologic evidence of tumor recurrence (18–21).
Glioblastomas, Metastatic Tumors, and Malignant Lymphomas
These tumors are sometimes difficult to differentiate from each other when only conventional MR imaging studies are available because they usually manifest as enhanced masses. The ADC of glioblastomas is reportedly lower than that of metastatic tumors (2). However, our results indicate that the ADC is not useful for differentiating between glioblastomas and metastatic tumors.
Reportedly, malignant lymphomas manifested lower ADCs than did high-grade astrocytomas (15). We restricted our analysis to glioblastomas and excluded anaplastic astrocytomas because the latter sometimes show weak or no enhancement. We found that malignant lymphomas manifested lower ADCs than those of glioblastomas and metastatic tumors. Our results suggest that preoperative evaluation of the ADC may make it possible to obtain a differential diagnosis of malignant lymphoma.
Meningiomas and Schwannomas
According to Kono et al (14), the ADC is not indicative of the histologic subtype of meningiomas. Others (3) reported that atypical (WHO grade 2) and anaplastic (WHO grade 3) meningiomas exhibited lower ADC than that of benign meningioma subtypes. We found that the ADC may not be predictive of the degree of malignancy in meningiomas or of their histologic subtype. Studies in large populations are needed to draw definitive conclusions.
To our knowledge, the ADCs of meningiomas and schwannomas—tumors that are sometimes difficult to differentiate—have not been compared to date. We found that the ADC of schwannomas was significantly higher than that of meningiomas. Histologically, schwannomas comprise Antoni type A and type B neurinomas, and their higher ADC may reflect the lower cell density of Antoni type B neurinomas. We posit that preoperative evaluation of the ADC of tumors at the cerebellopontine angle and the middle cranial fossa may provide useful information for differentiating between meningiomas and schwannomas. Some meningioma subtypes, however, such as microcystic meningiomas, may have higher ADC values than those of schwannomas.
Tumors Developing Parasellar and Pineal Lesions and Other Analyses
The preoperative differentiation of craniopharyngiomas and pituitary adenomas can be difficult. The squamous-papillary type of craniopharyngioma may contain only solid components, while pituitary adenomas may be complicated by cystic formation. In addition, it can be difficult to differentiate sellar meningiomas and pituitary adenomas. We found that the ADC is a useful parameter for differentiating between craniopharyngiomas and pituitary adenomas and between craniopharyngioma and meningioma but not for differentiating between pituitary adenomas and meningiomas.
Since both epidermoid tumors and chordomas exhibit high signal intensity on DW images, probably as a result of the T2 shine-through effects, which contribute to the DW images, the differentiation of these two tumors can be difficult. In our study, the ADC of epidermoid tumors was lower than that of chordomas. When compared with the ADC of normal brain, the ADC of epidermoids is somewhat higher, and that of chordomas is much higher (22,23). We posit that determination of the ADC may aid in differentiating between epidermoid tumors and chordomas. Since we cannot rule out that our single-shot echo-planar imaging-based diffusion techniques resulted in degradation because of the diamagnetic susceptibility effects of the skull base, we are in the process of performing fast spin-echo–based diffusion and ADC studies.
Large patient populations must be studied to evaluate the applicability of the ADC for obtaining a differential diagnosis of germ cell tumors, craniopharyngiomas, pituitary adenomas, epidermoid tumors, and chordomas.
Our study has some limitations. First, our results were compromised by the partial volume effect. Our section thickness at DW MR imaging was 7.5 mm, and thinner section thickness could reduce the partial volume effect. Second, our results may be affected by the spatial distortion of DW images. Multishot echo-planar imaging–based or fast spin-echo–based diffusion techniques will reduce this spatial distortion. In addition, studies with repetition times longer than the 1600 msec we used must be performed to obtain unequivocal results. Third, nonenhanced and enhanced tumors and grade 3 glioneuronal tumors must be addressed; however, only a few cases were available for analysis. Similarly, our series included only a few patients with rare tumors. Further studies are necessary to determine the usefulness of the ADC for differentiating these tumors.
Despite these caveats, we conclude that the ADC is useful for the differentiation of some human brain tumors. Our study revealed a good inverse correlation between ADC and WHO grade 2–4 astrocytic tumors. The ADC of malignant lymphomas was lower than that of glioblastomas and metastatic tumors. Interestingly, the ADC of DNT was higher than that of any other glioneuronal tumors. The ADC of medulloblastomas, PNETs, and pineoblastomas was lower than that of ependymomas; it was lower in meningiomas than in schwannomas, lower in pituitary adenomas than in craniopharyngiomas, and lower in epidermoid tumors than in chordomas. Although the preoperative assessment of the ADC can be useful for the differentiation of some tumors, because it overlaps in some tumor types, additional evaluation parameters are needed for unequivocal differentiation among different kinds of human brain tumors.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, F.Y., K.K.; study concepts, all authors; study design, F.Y., K.K., K.A.; literature research, F.Y., A.T., R.H., H.Y., Y.I., K.Y., T.S., M.A.T.; clinical studies, F.Y., K. Sugiyama, J.T., S.H., Y.K.; data acquisition, F.Y., K. Sugiyama, J.T., S.H., Y.K.; data analysis/interpretation, F.Y., J.T., Y.K.; statistical analysis, M.O., K. Satoh, F.Y.; manuscript preparation, F.Y.; manuscript definition of intellectual content, F.Y., K.K.; manuscript editing, revision/review, and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Murakami, T. Hirai, T. Sugahara, H. Fukuoka, R. Toya, S. Nishimura, M. Kitajima, T. Okuda, H. Nakamura, N. Oya, et al. Grading Astrocytic Tumors by Using Apparent Diffusion Coefficient Parameters: Superiority of a One- versus Two-Parameter Pilot Method Radiology, June 1, 2009; 251(3): 838 - 845. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E.J. Lee, S.K. Lee, R. Agid, J.M. Bae, A. Keller, and K. terBrugge Preoperative Grading of Presumptive Low-Grade Astrocytomas on MR Imaging: Diagnostic Value of Minimum Apparent Diffusion Coefficient AJNR Am. J. Neuroradiol., November 1, 2008; 29(10): 1872 - 1877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-H. Toh, M. Castillo, A.M.-C. Wong, K.-C. Wei, H.-F. Wong, S.-H. Ng, and Y.-L. Wan Differentiation Between Classic and Atypical Meningiomas with Use of Diffusion Tensor Imaging AJNR Am. J. Neuroradiol., October 1, 2008; 29(9): 1630 - 1635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V.A. Nagar, J.R. Ye, W.H. Ng, Y.H. Chan, F. Hui, C.K. Lee, and C.C.T. Lim Diffusion-Weighted MR Imaging: Diagnosing Atypical or Malignant Meningiomas and Detecting Tumor Dedifferentiation AJNR Am. J. Neuroradiol., June 1, 2008; 29(6): 1147 - 1152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.S. Seo, K.-H. Chang, D.G. Na, B.J. Kwon, and D.H. Lee High b-Value Diffusion (b = 3000 s/mm2) MR Imaging in Cerebral Gliomas at 3T: Visual and Quantitative Comparisons with b = 1000 s/mm2 AJNR Am. J. Neuroradiol., March 1, 2008; 29(3): 458 - 463. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-H. Toh, M. Castillo, A.M.-C. Wong, K.-C. Wei, H.-F. Wong, S.-H. Ng, and Y.-L. Wan Primary Cerebral Lymphoma and Glioblastoma Multiforme: Differences in Diffusion Characteristics Evaluated with Diffusion Tensor Imaging AJNR Am. J. Neuroradiol., March 1, 2008; 29(3): 471 - 475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B Hakyemez, N Yildirim, O Taskapilioglu, C Erdogan, S Aker, S Yilmazlar, and M Parlak Intracranial myeloid sarcoma: conventional and advanced MRI findings Br. J. Radiol., June 1, 2007; 80(954): e109 - e112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Murakami, T. Sugahara, H. Nakamura, T. Hirai, M. Kitajima, Y. Hayashida, Y. Baba, N. Oya, J.-i. Kuratsu, and Y. Yamashita Malignant Supratentorial Astrocytoma Treated with Postoperative Radiation Therapy: Prognostic Value of Pretreatment Quantitative Diffusion-weighted MR Imaging Radiology, May 1, 2007; 243(2): 493 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Rumboldt, D.L.A. Camacho, D. Lake, C.T. Welsh, and M. Castillo Apparent Diffusion Coefficients for Differentiation of Cerebellar Tumors in Children AJNR Am. J. Neuroradiol., June 1, 2006; 27(6): 1362 - 1369. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
