| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (E.M.H., A.E.H., G.M.I., W.Y.H., D.C.K., J.S.B., B.T., V.S.L., G.A.K.) and Pathology (A.B.W.), New York University Medical Center, 560 First Ave, Suite HW 202, New York, NY 10016. Received April 1, 2005; revision requested May 31; revision received June 22; accepted July 11. Address correspondence to: E.M.H. (e-mail: hechte01{at}med.nyu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Materials and Methods: A waiver of informed consent and institutional review board approval for this retrospective study were granted. The study was HIPAA compliant. Twenty-eight men (mean age, 49 years; range, 23–70 years) and 10 women (mean age, 53 years; range, 42–72 years) with cirrhosis underwent T2-weighted and contrast-enhanced T1-weighted MR imaging at 1.5 T within 90 days of liver transplantation. Three readers reviewed the T1-weighted images alone and then the T2-weighted and T1-weighted images together. Lesion detection, characterization, and reader confidence levels were recorded.
Results: At liver explantation, 57 lesions were present in 18 patients: 19 HCCs, 33 dysplastic nodules, and five cysts. Contrast-enhanced T1-weighted imaging depicted 13 of 19 HCCs with an overall sensitivity of 68.4% (13 of 19) and specificity of 65.7% (23 of 35). The sensitivity and specificity for detection of dysplastic nodules (sensitivity, 9%; specificity, 68.4%) and HCCs (sensitivity, 68.4%; specificity, 65.7%) were nearly identical for T1-weighted images read alone or read with T2-weighted images. The only difference was the specificity for T1-weighted images read alone (65.7%) and read with T2-weighted images (62.9%). The addition of T2-weighted images altered the diagnosis in one of 90 (1.1%) cases and provided an increase in diagnostic confidence in four of 258 (1.6%) cases for independent readers and three of 90 (3.3%) cases at consensus reading.
Conclusion: Contrast-enhanced T1-weighted imaging can be used as a stand-alone sequence for the diagnosis of HCC in patients with cirrhosis prior to liver transplantation.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Magnetic resonance (MR) imaging plays a prominent role in the evaluation of cirrhosis and screening for early HCC (7–9). Most MR imaging protocols rely on both T2-weighted imaging and multiphase dynamic gadolinium-enhanced imaging to depict and characterize tumors. HCCs, however, demonstrate variable signal intensity on T2-weighted images, and many small (<2 cm in diameter) HCCs are occult on T2-weighted MR images (10).
Several recent studies with surgical correlation have shown that dynamic two-dimensional or three-dimensional (3D) gadolinium-enhanced MR imaging is superior to T2-weighted MR imaging for diagnosing focal lesions in the cirrhotic liver (11–15). However, because these studies relied on surgical correlation only, without whole liver explant correlation, the true accuracy of different MR imaging methods could not be determined. Specifically, sensitivities for HCC may have been overestimated, and dysplastic nodules could not be evaluated. Furthermore, the rate of false-positive findings for HCC at dynamic gadolinium-enhanced MR imaging could not be accurately assessed without whole liver radiologic-pathologic correlation. Thus, the purpose of our study was to retrospectively assess the sensitivity, specificity, and reader confidence in diagnosing HCC in the cirrhotic liver at contrast material–enhanced T1-weighted MR imaging alone and to determine if the addition of T2-weighted MR imaging improves sensitivity, specificity, and reader confidence assessed at contrast-enhanced T1-weighted imaging alone by using pathologic examinations as the reference standard.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our institutional liver transplantation database was retrospectively cross-referenced with our MR imaging database to identify all patients with cirrhosis who underwent both transplantation and contrast-enhanced T1-weighted MR imaging within a 90-day period from December 1999 to June 2003. The search yielded 50 patients, 12 of whom were excluded because of previous chemoembolization. Of the 38 patients included in the final cohort, 28 were men (mean age, 49 years; range, 23–70 years), and 10 were women (mean age, 53 years; range, 42–72 years). The 38 patients had a mean age of 54 years (range, 23–72 years). The cause of cirrhosis was either hepatitis C (n = 18), primary biliary cirrhosis (n = 4), hepatitis B (n = 4), ethanol abuse (n = 3), cryptogenic cirrhosis (n = 4), hemochromatosis (n = 1), primary biliary cirrhosis and hemachromatosis (n = 1), ethanol abuse and hepatitis C (n = 1), autoimmune hepatitis (n = 1), or Wilson disease (n = 1).
MR Imaging
MR imaging was performed at 1.5 T (Vision or Symphony; Siemens Medical Systems, Erlangen, Germany) with a torso phased-array coil. All patients were imaged with the following pulse sequences: transverse breath-hold T2-weighted echo train turbo short inversion time inversion-recovery (turbo STIR) and transverse 3D fat-suppressed T1-weighted spoiled gradient-echo acquisition (16).
Parameters for the T2-weighted sequences were as follows: repetition time, 3600–5074 msec; (effective) echo time, 76–86 msec (3600–5074/76–86); refocusing flip angle, 150°–180°; matrix, 99–132 x 256; rectangular field of view optimized to the patient's body habitus, 300–400 mm; section thickness, 8 mm; intersection gap, 2 mm; echo train length, 33; inversion time, 150–165 msec; and bandwidth, 325 Hz/pixel. Two interleaved breath-hold acquisitions with eight to 10 sections were required for sufficient anatomic coverage. Parameters for the T1-weighted acquisitions were as follows: repetition time, 3.3–4.5 msec; echo time, 1.4–1.9 msec (3.3–4.5/1.4–1.9); flip angle, 12°; matrix, 128–192 interpolated to 256 x 256; rectangular field of view optimized to the patient's body habitus, 300–400 mm; interpolated section thickness, 2–3 mm; slab thickness, 160–200 mm to ensure full coverage of the liver; and bandwidth, 488–490 Hz/pixel. Phase encoding was acquired in a sequential manner.
The T1-weighted sequence was performed once before and three times after intravenous administration of 19 mL of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) followed by a 20-mL saline flush at a rate of 2 mL/sec by using a power injector (Spectris; Medrad, Pittsburgh, Pa). To determine the timing for the first contrast-enhanced acquisition (hepatic arterial phase), a 1-mL test bolus of contrast material was administered and the circulation time (time to peak arterial enhancement) was set as the acquisition delay time (17,18). The second acquisition (portal venous phase) was performed 60 seconds after the administration of contrast material, and the final acquisition (equilibrium phase) was performed 3 minutes after the administration of contrast material. All acquisitions were performed during suspended respiration at end expiration to optimize image coregistration for subtraction algorithms. Subtraction of unenhanced images from contrast-enhanced images was performed to improve the detection of lesion enhancement.
Image Analysis
Three radiologists (E.M.H., D.C.K., W.Y.H.), each with 2 years of experience in MR image evaluation, were blinded to pathologic results and retrospectively and independently reviewed all 38 MR studies on a commercially available workstation (Syngo; Siemens Medical Systems). Initially, the unenhanced, contrast-enhanced, and subtracted T1-weighted data sets were evaluated. Readers were asked to note if there were any substantial artifacts and to indicate if images were uninterpretable. For interpretable images, all readers were asked to measure and document all lesions on liver maps, to characterize each lesion according to signal intensity and enhancement characteristics as described in the next section, and to ascribe a confidence score to each diagnosis. The confidence scale was based on whether each lesion represented HCC with scores ranging from 1 to 5 (1, not HCC; 2, probably not HCC; 3, indeterminate; 4, probably HCC; 5, definitely HCC). For statistical purposes, a lesion diagnosed as HCC by the readers was determined to be a positive finding only if a confidence score of 4 or 5 was given; any score less than that was considered a negative finding for HCC.
Later the same day, the T2-weighted images were reviewed with the T1-weighted images. Lesions and reader confidence levels were recorded. Up to 10 lesions were recorded for each patient. If more than 10 lesions were seen (n = 1), readers indicated that there were innumerable lesions but only the largest lesions were recorded.
One investigator (A.E.H.), a 3rd-year radiology resident not involved in lesion detection or characterization, compiled interpretations of all the readers and noted any disagreements, which were then resolved by a consensus reading that included all readers. This same investigator reviewed and recorded the signal intensity and enhancement characteristics of all lesions. For cases in which HCCs were seen at pathologic examination and not by the readers, images were reviewed to determine if HCC could be seen in retrospect.
Lesion Characterization
Readers were asked to categorize lesions according to signal intensity characteristics and enhancement pattern into one of five categories: HCC, dysplastic nodule, hepatic arterial pseudolesion, hemangioma, and cyst or biliary hamartoma.
HCC was diagnosed if the lesion fulfilled any two of the four following criteria: moderate hyperintensity on T2-weighted images when compared with surrounding liver parenchyma; hepatic arterial enhancement; portal venous or equilibrium phase washout; and capsule at unenhanced or delayed contrast-enhanced imaging (10,19). A dysplastic nodule was diagnosed if the lesion fulfilled any of the following criteria: isointensity or hypointensity on T2-weighted images, hyperintensity on unenhanced T1-weighted images, lack of hepatic arterial phase enhancement, and size substantially larger (twice the size) than background regenerative nodules (7). Regenerative nodules were not recorded.
A hepatic arterial pseudolesion was diagnosed if the lesion was identified only on the hepatic arterial phase T1-weighted image; was located within 2 cm of the capsule, adjacent to the gallbladder fossa, or posteriorly in liver segment IV; and was not round (20,21). A hemangioma was diagnosed if the lesion fulfilled at least three of the four following criteria: isointensity on T2-weighted images compared with cerebrospinal fluid, enhancement in a peripheral nodular pattern, flash filling during the hepatic arterial phase, and signal intensity during portal venous phase similar to that of the portal or hepatic vein (22–24). A cyst or biliary hamartoma was characterized by T1 hypointensity similar to that of cerebrospinal fluid, T2 hyperintensity equal to cerebrospinal fluid, and a lack of enhancement following administration of contrast material (25,26).
Pathologic Analysis
All 38 explanted livers were sectioned into 5–8-mm contiguous slices in either the coronal or transverse plane by one of two pathologists (including A.B.W., with 18 years of experience). The pathologists were blinded to the preexplantation imaging results. Dysplastic nodules and HCCs were identified grossly as those that were distinct from surrounding regenerative nodules in terms of size, texture, color, or degree of bulging beyond the cut surface of the liver (27). Livers were photographed, and all lesions other than ordinary regenerative nodules were sampled for histologic examination.
By using the diagnostic criteria of the International Working Party's Terminology of Nodular Hepatocellular Lesions, the routine hematoxylin-eosin–stained slices from the nodules were classified as follows: regenerative nodule; dysplastic nodule, low grade; dysplastic nodule, high grade; small HCC (<2 cm); or HCC (>2 cm) (27). There were no absolute size criteria, as described by the International Working Party, to diagnose or distinguish regenerative nodules from dysplastic nodules.
Statistical Analysis
All statistical computations were performed by using statistical software (SAS system for Windows, version 9.0; SAS Institute, Cary, NC), and results were statistically significant only if associated with a two-sided P value less than .05. Cohen 
was used to assess interreader agreement with respect to diagnoses determined by each reader on the basis of T1-weighted images alone and T1-weighted images with the addition of T2-weighted images. The reported is an average of the values computed for the three readers over the entire set of all lesions. Generalized estimating equations based on a binary logistic regression model were used to compare T1-weighted images alone and with the addition of T2-weighted images in terms of diagnostic accuracy (ie, the proportion of times diagnoses were concordant with pathologic findings) and to provide confidence bounds for the proportion of times the combination of both types of images was expected to lead to the same diagnosis and the proportion of times the combination of both types of images led to an increase in reader confidence. The logistic regression model included patient sex and reader identification as fixed classification factors and lesion size (as determined at pathologic examination) as a fixed numeric factor (when assessing sensitivity only). The covariance structure was modeled by assuming observations to be independent or correlated when obtained from the same patient or different patients, respectively, with the correlation strongest when observations were associated with the same patient.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MR Imaging Sensitivity and Specificity
For the detection of HCC with whole liver explant correlation, contrast-enhanced T1-weighted imaging had a sensitivity of 68.4% (13 of 19) and specificity of 65.7% (23 of 35), while the combination of contrast-enhanced T1-weighted and T2-weighted imaging yielded a sensitivity of 68.4% (13 of 19) and specificity of 62.9% (22 of 35). For HCCs larger than 1 cm in diameter, sensitivity was 86.7% (13 of 15). For dysplastic nodules, the sensitivity was 9% (three of 33) and specificity was 68.4% (26 of 38) for contrast-enhanced T1-weighted imaging alone and for both techniques in combination.
There was a significant increase in sensitivity (P < .001) for the detection of dysplastic nodules, HCCs, or both as the lesion size at pathologic examination increased. The smallest pathologically confirmed nodule classified as a dysplastic nodule by at least one reader with either contrast-enhanced T1-weighted images alone or in conjunction with T2-weighted images was 0.8 cm. The sensitivity for dysplastic nodules 0.8 cm or larger was 19.6% for individual reader opinion and 23.5% for the consensus opinion of three readers. The sensitivity for HCCs smaller than 1 cm was 25% for both individual reader and consensus opinions.
Imaging Features of HCC and Dysplastic Nodule
Signal intensity and enhancement patterns of the HCCs varied (Table). Of 14 HCCs seen on T2-weighted images, three were hypointense, five were isointense, and six were hyperintense (Fig 1). Signal intensity on unenhanced T1-weighted images varied from hypointense (n = 2) to isointense (n = 1) to hyperintense (n = 11). Eleven of 19 (58%) HCCs were more apparent on hepatic arterial phase T1-weighted images; one was more apparent on the portal venous phase T1-weighted image, and two were most apparent on equilibrium phase T1-weighted images as hypointense lesions.
|
|
|
|
|
Contrast-enhanced T1-weighted Imaging Alone versus in Combination with T2-weighted Imaging
Overall, 90 lesions were detected at T1-weighted imaging; only 23 of these lesions had a pathologic correlate (Fig 2). Contrast-enhanced T1-weighted imaging depicted 14 of 19 (74%) pathologically proved HCCs, although one HCC was given a confidence score of only 3 and was not considered a true-positive finding (mean size, 2.8 cm; range, 0.5–8.6 cm). Four of 33 (12%) pathologically proved dysplastic nodules were depicted (mean size, 0.9 cm; range, 0.8–1.0 cm), and five of five pathologically proved cysts were depicted (mean size, 0.5; range, 0.4–0.6 cm).
|
|
|
|
|
|
coefficients were each estimated as 0.99 for both the consensus opinions and the results provided by each of the individual readers. Overall, the data indicate that there is at least 95% confidence that the addition of information from T2-weighted images to that of T1-weighted images will result in an alteration of the diagnosis less than 3% of the time. In terms of confidence scores, of the 14 pathologically confirmed HCCs seen on contrast-enhanced T1-weighted images, 11 lesions were given a confidence score of 5, two were given a confidence score of 4, and one was given a confidence score of 3, indeterminate, and not considered a true-positive finding. In one case, the addition of T2-weighted images led to a change in the level of confidence in diagnosis of HCC from 4 to 5 in the consensus reading.
Of the four pathologically confirmed dysplastic nodules seen on contrast-enhanced T1-weighted images, three were given a confidence score of 2, and one was given a score of 3. The addition of T2-weighted images did not alter the level of confidence in the diagnosis of dysplastic nodules in the consensus reading.
Overall, the readers reported the exact same level of confidence when relying on contrast-enhanced T1-weighted images alone or with the addition of T2-weighted images 98.4% of the time. The addition of T2-weighted images increased diagnostic confidence in four of 258 confidence scores (1.6%) for independent readers and three of 90 confidence scores (3.3%) for the consensus reading. On four of 258 (1.6%) occasions, reader confidence was exactly one unit higher for the diagnoses from contrast-enhanced T1-weighted images and T2-weighted images when compared with those determined from contrast-enhanced T1-weighted images alone. Reader confidence was never higher for diagnoses based on contrast-enhanced T1-weighted images alone. On no occasion did the confidence scores provided by the same reader for the same lesion differ by more than one unit. There was at least 95% confidence that the addition of T2-weighted MR images would lead to an increase in diagnostic confidence less than 5% of the time when readers provided independent opinions and no more than 8.5% of the time when three readers provided a consensus opinion.
Lesions Detected Only at MR Imaging
Sixty-seven of 90 (74%) lesions seen on the contrast-enhanced T1-weighted images had no pathologic correlate. Nineteen of these 67 (28%) lesions were thought to represent HCC with confidence scores of 5 (n = 2), 4 (n = 10), and 3 (n = 7) on the basis of contrast-enhanced T1-weighted images alone (Fig 1). In two cases, the addition of T2-weighted to T1-weighted images changed the confidence score at consensus reading from 3 to 4 (n = 1) and 4 to 3 (n = 1). In one case, the addition of the T2-weighted images led to a change in diagnosis of a pseudolesion at hepatic arterial phase imaging (confidence score of 2) to HCC (confidence score of 3) (Fig 2).
Thirteen of 67 (19%) lesions were thought to represent dysplastic nodules; 12 were given a confidence score of 2, and one was given a score of 3. The addition of T2-weighted images did not alter the confidence scores at consensus reading in these cases.
The remaining 35 lesions were thought to represent pseudolesions on hepatic arterial phase T1-weighted images and were given confidence scores of 1 (n = 1), 2 (n = 33), or 3 (n = 1). The addition of T2-weighted images did not alter confidence scores in diagnosis of these lesions.
Analysis on a per Patient Basis
A finding was classified as a true-positive if at least one lesion was classified as HCC or dysplastic nodule at pathologic examination. There were too few patients for a meaningful breakdown of per patient specificity by dysplastic nodules and HCCs separately.
Overall, there were 18 subjects with at least one dysplastic nodule or HCC at pathologic examination. Twelve of 18 (68%) patients had a dysplastic nodule or HCC detected on T1-weighted images alone and with the addition of T2-weighted images. There was no significant difference (P > .99) between contrast-enhanced T1-weighted images read alone and contrast-enhanced T1-weighted images read with T2-weighted images in terms of sensitivity for the identification of subjects with HCC.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Our findings that T2-weighted imaging is insensitive for identifying HCC are consistent with findings in previous reports. In a study by Hussain et al, only 14 of 30 (47%) HCCs were detected as hyperintense masses on T2-weighted images (15). Fujita et al demonstrated that 21.6%–32.4% of HCCs were not hyperintense on images from T2-weighted breath-hold fast spin-echo sequences with both short ([effective] echo time, 93 msec) and long ([effective] echo time, 138 msec) sequences (13).
In our study, the addition of T2-weighted MR imaging did not improve the accuracy or confidence in diagnosis of HCCs or dysplastic nodules when compared with that of contrast-enhanced T1-weighted imaging alone. The addition of T2-weighted imaging led to a change in reader confidence in three cases (of which only one was a pathologically proved HCC) and incorrectly led to a change of diagnosis in one case (from a pseudolesion at hepatic arterial phase T1-weighted imaging to an HCC). For all other liver masses identified on the T2-weighted images, there was no change in reader confidence or diagnosis when compared with those from contrast-enhanced T1-weighted images alone. Our results with whole liver explant correlation are in agreement with those of previous studies that assessed the utility of T2-weighted imaging in detection and characterization of HCC. Fujita et al showed dynamic gadolinium-enhanced T1-weighted imaging had 100% sensitivity for detection of HCC when compared with T2-weighted imaging, which had a sensitivity of only 68%–78% (13). On the basis of results of multireader receiver operating characteristic analysis, Hussain et al demonstrated that there was no significant difference in HCC detection or characterization with the addition of T2-weighted images (15).
Our study has recognized limitations. The T2-weighted sequence we used, a transverse breath-hold echo train turbo STIR sequence, is one that is commonly implemented in clinical protocols but that may have intrinsic shortcomings. Echo train imaging, although widely used clinically because of its improved efficiency, is known to result in decreased image contrast compared with image contrast at conventional T2-weighted spin-echo imaging. Moreover, the additive T1 and T2 weighting achieved with the inversion-recovery sequence may impair detection of the HCC because of short T1 relaxation times; in our study, 11 of 19 (58%) HCCs appeared hyperintense on T1-weighted images, but only four of the 11 were hyperintense on T2-weighted images. We prefer the inversion-recovery method of fat suppression over frequency-selective fat saturation, however, because of the greater robustness and insensitivity to magnetic field inhomogeneity of the inversion-recovery method. In addition, the T2-weighted imaging was performed with a conventional section thickness of 8 mm (with 2-mm gap) compared with thin-section (interpolated 2–3 mm) contrast-enhanced T1-weighted imaging, which represents a bias toward the latter technique. Finally, our results cannot be extrapolated to include centers that use both ferumoxide-enhanced T2-weighted and gadolinium-enhanced 3D T1-weighted MR imaging (34).
An additional consideration in evaluating our results is the accuracy of pathologic correlation. Pathologic specimens were examined by using 5–8-mm slices. It is possible that small HCCs seen on MR images may have been overlooked at pathologic examination. Review of the pathologic specimens with attention to the smaller lesions suspected of being HCC that had no pathologic correlate was not performed for all lesions.
Our study demonstrates that T2-weighted imaging added to T1-weighted imaging does not improve the detection of HCC or increase reader confidence in characterizing focal masses in the cirrhotic liver when compared with contrast-enhanced T1-weighted imaging performed with a test bolus for accurate timing of the hepatic arterial phase acquisition. While the sensitivity of MR imaging for the detection of HCC larger than 1 cm was 86.7%, our results support the growing body of literature that imaging is insensitive for lesions smaller than 1 cm (25% in our study) and dysplastic nodules. The clinical importance of this limitation is mitigated by the high survival rate of subjects with HCC smaller than 3–5 cm after undergoing transplantation (4). Nonetheless, given the rapid doubling time of some HCCs (35–37), the development of improved imaging methods for the detection of small HCC is an area of active research.
|
FOOTNOTES |
|---|
Abbreviations: HCC = hepatocellular carcinoma • STIR = short inversion time inversion recovery • 3D = three dimensional
2 Current address: Department of Radiology, Georgetown Hospital, Washington, DC. 
3 Current address: East River Medical Imaging, New York, NY.
4 Current address: The Valley Hospital, Ridgewood, NJ.
Author contributions: Guarantor of integrity of entire study, E.M.H., B.T.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, E.M.H., G.M.I., B.T.; clinical studies, E.M.H., A.E.H., G.M.I., W.Y.H., D.C.K., A.B.W., V.S.L., G.A.K.; statistical analysis, E.M.H., J.S.B., B.T.; and manuscript editing, E.M.H., A.E.H., G.M.I., W.Y.H., D.C.K., J.S.B., B.T., V.S.L., G.A.K.
Authors stated no financial relationship to disclose.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. H. Choi, J. M. Lee, N. C. Yu, K.-S. Suh, J.-J. Jang, S. H. Kim, and B. I. Choi Hepatocellular Carcinoma in Liver Transplantation Candidates: Detection with Gadobenate Dimeglumine-Enhanced MRI Am. J. Roentgenol., August 1, 2008; 191(2): 529 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Willatt, H. K. Hussain, S. Adusumilli, and J. A. Marrero MR Imaging of Hepatocellular Carcinoma in the Cirrhotic Liver: Challenges and Controversies Radiology, May 1, 2008; 247(2): 311 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Parikh, S. J. Drew, V. S. Lee, S. Wong, E. M. Hecht, J. S. Babb, and B. Taouli Focal Liver Lesion Detection and Characterization with Diffusion-weighted MR Imaging: Comparison with Standard Breath-hold T2-weighted Imaging Radiology, March 1, 2008; 246(3): 812 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Heilmaier, R. Sutter, A. M. Lutz, B. Seifert, D. Weishaupt, B. Marincek, and J. K. Willmann Mapping of Hepatic Vascular Anatomy: Dynamic Contrast-enhanced Parallel MR Imaging Compared with 64 Detector Row CT Radiology, December 1, 2007; 245(3): 872 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Lauenstein, K. Salman, R. Morreira, T. Heffron, J. R. Spivey, E. Martinez, P. Sharma, and D. R. Martin Gadolinium-Enhanced MRI for Tumor Surveillance Before Liver Transplantation: Center-Based Experience Am. J. Roentgenol., September 1, 2007; 189(3): 663 - 670. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
