| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Department of Radiology, New York University Medical Center, 560 First Ave, New York, NY 10016 (B.A.B., J.E.J., J.S.B.); Department of Radiology, Scottsdale Memorial Hospitals, Ariz (D.D.M.); and Department of Radiology, University of Pennsylvania Medical Center, Philadelphia (D.P.C.). From the 2000 RSNA scientific assembly. Received May 15, 2001; revision requested June 26; final revision received May 29, 2002; accepted June 5. Address correspondence to B.A.B. (e-mail: bernard.birnbaum@med.nyu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: A customized anthropomorphic phantom was designed to accept 40-, 140-, and 240-HU renal inserts containing intrarenal and exophytic 7-, 10-, and 15-mm cysts. Each phantom and insert were scanned with five different helical CT scanners by using 1.0–1.5-mm, 2.50–3.75-mm, 5.0-mm, 7.0–8.0-mm, and 10.0-mm section collimation. Means and SDs of CT number measurements were obtained for each cyst within each variably "enhanced" renal insert. Mixed-model analysis of variance accommodating heteroscedasticity of data was used to assess the effect of scanner type, section collimation, and cyst diameter on cyst attenuation.
RESULTS: Pseudoenhancement (range, 10.3–28.3 HU), observed by using effective section collimation equal to or less than 50% of cyst diameter, occurred in 34 (38%) of 90 intrarenal cyst measurements. Pseudoenhancement was observed with all five CT scanners, though the magnitude of the effect was nonuniform. Significant interactions were noted between renal cyst diameter, background renal attenuation, and CT scanner type in terms of their effects on cyst attenuation. No appreciable pseudoenhancement was observed with exophytic cysts.
CONCLUSION: Pseudoenhancement is maximal when small (
1.5-cm) intrarenal cysts are scanned during maximal levels of renal parenchymal enhancement. The magnitude of this effect varies with scanner type but may be large enough to prevent accurate lesion characterization, despite use of a thin-section helical CT data acquisition technique.
© RSNA, 2002
Index terms: Computed tomography (CT), experimental studies, 81.12111, 81.12112 • Experimental study • Kidney, CT, 81.12111, 81.12112 • Kidney, cysts, 81.3111 • Phantoms
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The pseudoenhancement effect has been confirmed to date in simple CT phantom studies in which fluid attenuation "cysts" have been scanned after being suspended in iodine baths of varying concentration and background attenuation (3,4). By scanning cylindric cysts free of partial-volume effects, investigators have shown that the effect is not primarily caused by volume averaging (3). Results of computer-simulated data analysis suggest that pseudoenhancement may result from inadequate CT algorithm correction for beam hardening; however, this is only a working hypothesis at this point (3).
The purpose of our study was to determine the effects of cyst diameter and location (intrarenal, exophytic), renal attenuation, section collimation, and interscanner variability on renal cyst pseudoenhancement in a phantom model.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
Each renal insert contained tissue-equivalent material of appropriate shape and mass-attenuation coefficients to simulate simple, round intrarenal and exophytic renal cysts (0 HU, 1% linear attenuation tolerance). The cysts measured 7, 10, and 15 mm in diameter and were arranged in three "cyst pairs" (intrarenal and exophytic cysts of the same size) that were equally distributed along the z axis of each renal insert.
Imaging
The abdominal phantom was imaged with five helical CT scanners that were made available to us at three hospitals within our medical center. The CT systems included a single–detector row HiSpeed Advantage scanner (GE Medical Systems, Milwaukee, Wis), a single–detector row CTI scanner (GE Medical Systems), a multi–detector row LightSpeed QXI scanner (GE Medical Systems), a single–detector row Somatom Plus 4 scanner (Siemens Medical Systems, Iselin, NJ), and a single–detector row PQ 5000 scanner (Marconi Medical Systems, Cleveland, Ohio). All five scanners were calibrated according to the manufacturer’s specifications prior to data acquisition.
In each scanning session, the abdominal phantom was centered on the CT table to simulate normal anatomic positioning. Data acquisition sessions consisted of three groups of scan sequences to allow the 40-, 140-, and 240-HU renal inserts to be scanned individually within the phantom with a given set of scan parameters. Image acquisition and reconstruction parameters for the individual CT scanners were as follows: (a) HiSpeed Advantage and CTI scanners (section collimation, 1, 3, 5, 7, and 10 mm; helical pitch, 1:1; reconstruction interval, 50%; field of view (FOV), 33 and 32 cm, respectively; 120 kVp; 220 mAs; and scanning time, 1 second); (b) LightSpeed QXI scanner (section collimation, 1.25, 2.5, 3.75, 5.0, 7.5, and 10.0 mm; helical pitch, 3:1; reconstruction interval, 50%; FOV, 32 cm; 120 kVp; 220 mAs; and scanning time, 0.8 second); (c) Somatom Plus 4 scanner (section collimation, 1, 3, 5, 8, and 10 mm; helical pitch, 1:1; reconstruction interval, 50%; FOV, 35.6 cm; 120 kVp; 220 mAs; and scanning time, 1 second); and (d) PQ 5000 scanner (section collimation, 1.5, 3.0, 5.0, 8.0, and 10.0 mm; helical pitch, 1:1; reconstruction interval, 50%; FOV, 40 cm; 120 kVp; 225 mAs; and scanning time, 1 second). The FOV exceeded 32–33 cm at two hospital sites, where the Somatom Plus 4 and PQ 5000 systems were tested. These differences arose because the FOV parameters were not rigidly fixed and were subject to modification by the technologist who performed CT at these sites.
Data Collection
The mean attenuation value of each renal cyst was determined by drawing three overlapping circular regions of interest (ROIs) within each cyst lumen and then calculating a mean cyst attenuation measurement. ROIs were drawn by one investigator (D.D.M.) on the center section obtained through each cyst. The ROI areas were approximately 4 mm2 (7-mm cyst), 18 mm2 (10-mm cyst), and 54 mm2 (15-mm cyst). A total of 1,404 individual ROI measurements (468 mean cyst attenuation values) were obtained to sample each combination of cyst diameter and location, section collimation, and background renal attenuation for all five CT scanners. Cyst enhancement was determined at background renal attenuation levels of 140 and 240 HU by measuring the difference in mean cyst attenuation values between the 140- and 40-HU renal insert images and between the 240- and 40-HU renal insert images, respectively.
Data Analysis
Data analysis focused on those scanning conditions in which section collimation was equal to or less than 50% of cyst diameter to minimize or negate partial volume-averaging effects. The relationship between cyst diameter, renal attenuation, section collimation, scanner type, and cyst enhancement was investigated by performing mixed-model analysis of variance (ANOVA) with adjustment for inequality of variance across the levels of the factors under investigation. Specifically, a least-squares method was used to fit a general linear model to the enhancement data acquired at 140- and 240-HU background attenuation levels. The same underlying model was used to examine changes in the variability of the pseudoenhancement data across the levels of each factor and to assess the influence of the factors on the mean level of pseudoenhancement, while accommodating variance heterogeneity. Factors with a significance level (ie, P value) of less than .05 were considered to have a statistically significant effect on cyst enhancement. A Tukey-Kramer adjustment was applied to make pairwise comparisons among the levels of one factor at specific combinations of the levels of the other factors, while maintaining the familywise type I error rate at or below the 5% level. The analyses were conducted by using version 8.00 of the SAS System software (SAS Institute, Cary, NC).
We used ROIs that were less than 50% of cyst area to ensure that partial-volume effects were minimized in the experiment. To appreciate the statistical significance of this decision, we calculated representative standard error of the mean (SEM) measurements for different combinations of cyst diameter, ROI area, and section collimation. The SEM was calculated as follows: SEM = SD/
(N - 1), where N is the number of pixels in the ROI, and the SD and SEM are measured in Hounsfield units (2). The SDs used for this analysis were the aggregate mean SDs calculated for each section collimation parameter (see above). The number of pixels was calculated as follows: N = (ROI area)(reconstruction matrix size)/(reconstruction FOV)2, where ROI area is measured in square centimeters and the FOV is measured in centimeters. In these calculations, the ROI area (0.04–0.54 cm2) varied according to cyst diameter, the reconstruction matrix was 512 x 512, and the FOV was set to 35 cm, which represented the average FOV size used in our experiment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
Although pseudoenhancement of renal cysts was observed with all five CT scanners, scanner response and magnitude of the effect were nonuniform (Tables 1, 2). The effect appeared to be least pronounced with the Siemens Somatom Plus 4 scanner, in which intrarenal cysts increased in mean attenuation to a maximum of 15.5 HU. The maximum observed increase in mean CT attenuation values was 28.3 HU, noted with the Marconi PQ 5000 scanner. Intermediate maximum pseudoenhancement levels were noted with the GE CT scanners, in which maximum changes in mean CT attenuation values of 21.2, 24.1, and 21.7 HU were observed by using the HiSpeed Advantage, CTI, and QXI scanners, respectively. Mean intrarenal cyst enhancement data are presented in Table 2. Intrarenal cysts experienced a mean increase in attenuation of 6.5 HU ± 7.7 when the 240-HU renal insert was scanned with the Siemens Somatom Plus 4 system. This value represented the smallest artifactual increase in mean cyst attenuation observed among the five scanners studied at this level of background attenuation. Although intrarenal cysts demonstrated substantially less enhancement with the Siemens Somatom Plus 4 scanner at 240-HU background attenuation when compared with that of the other systems tested, these differences were not statistically significant (Fig 3).
|
|
|
At 140-HU background renal attenuation and 7-mm cyst diameter, significant differences existed between the CTI and HiSpeed Advantage, PQ 5000, and QXI scanners (P < .05). At 140-HU attenuation and 10-mm cyst diameter, no significant differences were observed between scanners. At 140-HU attenuation and 15-mm cyst diameter, significant differences existed between the CTI and HiSpeed Advantage scanners and the HiSpeed Advantage and QXI scanners (P < .05).
At 240-HU background renal attenuation and 7-mm cyst diameter, significant differences existed between the CTI and PQ 5000, QXI, and Somatom Plus 4 scanners and between the HiSpeed Advantage and Somatom Plus 4 scanners (P < .05). At 240-HU background renal attenuation and 10-mm cyst diameter, no significant differences were observed between scanners. At 240-HU background renal attenuation and 15-mm cyst diameter, significant differences existed between the PQ 5000 and CTI, HiSpeed Advantage, and QXI scanners (P < .05).
Evaluation of the intrarenal cyst enhancement data at 240-HU background attenuation revealed five cases (two at 1.0–1.5-mm collimation and three at 2.5–3.0-mm collimation) in which pseudoenhancement was not observed when 7-mm cysts were scanned but was observed when the same collimation was used to scan 10-mm (n = 4) or 15-mm (n = 4) cysts (Table 1). This set of paradoxical circumstances, in which pseudoenhancement was noted to preferentially affect larger rather than smaller cysts, was not seen at a background attenuation of 140 HU and was not observed with other collimation parameters. These five cases occurred by using the GE QXI (n = 1), Siemens Somatom Plus 4 (n = 2), and Marconi PQ 5000 (n = 2) scanners.
Evaluation of the exophytic cyst phantom attenuation data revealed the presence of volume-averaging effects as section collimation increased (all other factors held constant). No consistent or clinically relevant pseudoenhancement effects were observed when the exophytic cysts were scanned. Exophytic cyst enhancement data are presented in Table 3. Exophytic cysts were scanned with a section collimation equal to or less than 50% of cyst diameter in 90 (58%) of 156 enhancement measurements. An increase of more than 10 HU in cyst attenuation values was noted in only four (4%) of 90 cases. This scenario occurred while scanning 7-mm (n = 3) and 15-mm (n = 1) exophytic cysts with the GE CTI (n = 1), Siemens Somatom Plus 4 (n = 2), and Marconi PQ 5000 (n = 1) scanners. The absolute CT numbers in these four cases measured negative attenuation at all background levels tested (range, -0.5 HU to -29 HU).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This 10-HU enhancement criterion is based on anecdotal, albeit extensive, experience with the nonhelical CT technique. Maki et al (3) questioned the validity of this enhancement threshold. They reported difficulty in characterizing small intrarenal parenchymal cysts that appeared to exhibit pseudoenhancement of more than 10 HU if helical CT was performed during peak levels of renal enhancement. They validated the presence of the pseudoenhancement phenomenon in a polyethylene phantom study in which cylindric water-filled "cysts," designed to negate partial-volume effects, were suspended in iodine solutions of varying concentration. Their helical CT data revealed that cyst attenuation increased consistently with increasing background attenuation, that the effect was greater in cysts with smaller rather than larger diameters, and that cyst pseudoenhancement ranged from 10 to 28 HU as background attenuation increased by 180 HU (3).
Other investigators have now confirmed the pseudoenhancement effect in both in vitro (4) and in vivo studies (4,5). Results of these studies demonstrated that renal cyst pseudoenhancement is more likely as cyst diameter decreases and as cyst location becomes progressively intrarenal. Interestingly, the magnitude of pseudoenhancement varied in these investigations, which were performed by using different CT scanners. It is also noteworthy that Coulam et al (4) detected pseudoenhancement in cysts smaller than 2 cm in diameter, while Bae et al (5) observed this effect only in cysts measuring up to 1 cm. A possible explanation for these incongruous results is that the magnitude of pseudoenhancement varied in these experiments because the investigators used different CT systems that incorporated different CT reconstruction algorithms. To test this hypothesis, we designed a customized anthropomorphic body CT phantom that allowed us to determine how cyst diameter and location (intrarenal, exophytic), renal attenuation, section collimation, and interscanner variability affect renal cyst pseudoenhancement.
The results of our investigation confirmed the findings of previous phantom studies, which demonstrated that renal cyst attenuation might artifactually increase as background renal attenuation increases (3,4). In our experiment, pseudoenhancement of renal cysts was noted in seven cases in which background renal attenuation was 140 HU and in 27 cases in which background renal attenuation was 240 HU. Therefore, the effect was nearly four times as likely as renal enhancement doubled from 100 to 200 HU. These data support the observation that pseudoenhancement of renal cysts is most likely to be clinically apparent when helical renal CT scanning is performed during peak parenchymal enhancement, when renal attenuation is maximal (3).
By using an anatomic CT phantom constructed from tissue-equivalent materials, we demonstrated that purely exophytic renal cysts do not exhibit clinically important pseudoenhancement. These findings are consistent with clinical observations noted in recent in vivo studies (4,5). A limitation of our study is that we did not evaluate cysts that were partially exophytic in nature. Although it is possible that small partially exophytic renal cysts may demonstrate an artifactual increase in CT attenuation as renal attenuation increases, we did not test this specific hypothesis.
Our results corroborated the findings of earlier studies, which demonstrated that pseudoenhancement is more likely to occur as cyst diameter decreases (3–5). Nevertheless, we also identified five paradoxical cases in which 10- and 15-mm cysts demonstrated pseudoenhancement at a background attenuation of 240 HU, but the effect was not seen when 7-mm cysts were imaged with identical scanning conditions. This scanning anomaly was identified with three different CT scanners manufactured by three different CT vendors. We are unable to offer a plausible explanation for this observation, although it is possible that this may represent normal data scatter.
Pseudoenhancement of intrarenal cysts was observed with all five CT scanners tested; however, scanner response and effect magnitude were nonuniform. Maximum pseudoenhancement values occurred at 200-HU renal enhancement and ranged from 28.3 (Marconi PQ 5000 scanner) to 15.5 HU (Siemens Somatom Plus 4 scanner). Although no significant differences in overall renal cyst enhancement were noted among the five scanners studied, data analysis results nevertheless suggested that renal cysts scanned with the Siemens Somatom Plus 4 system demonstrated substantially less enhancement than that seen with the other CT scanners tested. This study finding is important and potentially clinically relevant and requires further investigation. It is currently hypothesized that renal cyst pseudoenhancement results from an inadequate CT algorithmic correction for beam hardening (3,5). It is possible that the CT algorithm incorporated into the Siemens Somatom Plus 4 system may be better suited to handle beam-hardening effects than are the CT algorithms used with the other scanners evaluated in this investigation. This may explain the discrepant in vivo study results reported by Coulam et al (4) and Bae et al (5). Since these investigators studied the pseudoenhancement effect by using different CT scanners that likely incorporated different proprietary CT algorithms produced by different vendors, it is of little surprise that the magnitude of pseudoenhancement varied in these studies. Differences in CT algorithmic handling of beam-hardening effects may also explain why some investigators have clinically observed renal cyst pseudoenhancement much more frequently than other investigators.
A potential limitation of our study design is that we did not specifically scan cylindric renal cysts. The advantage of doing so is that one is able to effectively eliminate partial volume-averaging effects by scanning through the center of such a fluid-filled structure. Maki et al (3) used this approach in a prior phantom study and demonstrated that renal cyst pseudoenhancement is clearly attributable to an artifact from something other than volume averaging alone. Our goal was to simulate clinically realistic scenarios in which helical CT is used to scan spherical renal cysts of varying diameter and location. We attempted to minimize or negate the effects of partial volume averaging by restricting our data analysis to those cases in which section collimation was equal to or less than 50% of cyst diameter and by ensuring that ROIs were drawn through the center of each cyst on images that were reconstructed with 50% overlap. It is noteworthy that pseudoenhancement of renal cysts occurred in 38% of cases in our phantom study, in which section collimation was no more than half the diameter of the cyst being evaluated. This emphasizes the fact that this phenomenon may preclude evaluation of small intrarenal cysts, despite use of optimized imaging parameters (4).
The absolute SD values for the cyst attenuation measurements obtained in the present study varied considerably. One would expect that as section collimation decreases, quantum noise and consequently the SD should increase, and this was confirmed by means of ANOVA, which held true for all scanners tested. When the mean SD data were used to calculate sample representative SEM, the SEM exceeded 3 HU in only a small minority of cases. It has been stated that the SEM is the appropriate statistic to use when comparing the mean values of two ROIs to determine if a mass is enhancing, and that an appropriate SEM value would be less than 3 HU (2). Our representative SEM measurements exceeded 3 HU when we measured the attenuation of 7-mm cysts, reflecting the fact that we used a small ROI of 4 mm2 to minimize or negate partial-volume averaging of renal parenchyma at the margins of these small cysts. This suggests that noise limitations may come into play when trying to measure enhancement of small renal cysts, since ROI size must be appropriately decreased to avoid volume averaging in these circumstances.
Additional limitations of our study include the fact that we did not evaluate the effects of beam hardening on all scanning parameters (milliampere seconds, peak kilovoltage, and reconstruction algorithm) and that we scanned our CT phantom with machines manufactured by only three CT vendors. It would have been advantageous to test additional CT algorithms that are incorporated into scanners produced by other manufacturers. Nevertheless, we were able to test five machines and demonstrate that variability in scanner response exists across different CT platforms. Because we evaluated only one multi–detector row helical CT scanner in our study, we cannot definitively state how this new technology will affect pseudoenhancement. If CT vendors use shared CT reconstruction algorithms between their single– and multi–detector row CT systems, however, we strongly believe that pseudoenhancement will be evident when using the new multi–detector row CT scanners.
Practical application: The results of the present study confirm that pseudoenhancement may be observed in simple, small intrarenal cysts (1.5 cm) when scanned at peak levels of renal enhancement, though scanner response and the magnitude of pseudoenhancement may vary considerably according to scanner type. Pseudoenhancement may not be apparent when scanning larger intrarenal (>1.5 cm) or exophytic cysts, despite high background renal attenuation levels.
The cause of pseudoenhancement is likely to be an inadequate CT algorithmic correction for beam hardening. We believe that this problem may be corrected by means of modification of clinically applicable CT reconstruction algorithms. Until such modification is made, radiologists should be aware that pseudoenhancement of renal cysts may occur, that the magnitude of cyst pseudoenhancement may vary according to scanner type, and that this problem may prevent accurate contrast material–enhanced helical CT characterization of small intrarenal lesions. Potential interim solutions to this problem include modification of contrast material dose according to patient weight to prevent "super enhancement" of the kidneys, increase of enhancement thresholds from the commonly used 10-HU level to 15–20 HU, and selective characterization of small intrarenal lesions (1.5 cm) by means of magnetic resonance (MR) imaging. In our own practice, we now use a 20-HU enhancement threshold when characterizing small (1.5 cm) intrarenal lesions in thin to moderate-sized patients who undergo imaging during the early nephrographic phase of renal enhancement. In those cases in which a well-defined intrarenal lesion demonstrates morphologic features strongly suggestive of a simple cyst but lesion enhancement exceeds 20 HU, correlation with MR imaging is advised to evaluate potential tumor neovascularity and to exclude the possibility of cyst pseudoenhancement.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, B.A.B., D.D.M., D.P.C., J.E.J.; study concepts and design, B.A.B., D.D.M., D.P.C.; literature research, B.A.B.; experimental studies, D.D.M., J.E.J.; data acquisition, D.D.M., J.E.J.; data analysis/interpretation, B.A.B., D.P.C., J.E.J.; statistical analysis, J.S.B., D.P.C.; manuscript preparation, B.A.B.; manuscript definition of intellectual content, B.A.B., D.D.M., D.P.C.; manuscript editing, revision/review, and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. J. Wang, F. V. Coakley, Y. Fu, B. N. Joe, S. Prevrhal, L. A. Landeras, E. M. Webb, and B. M. Yeh Renal Cyst Pseudoenhancement at Multidetector CT: What Are the Effects of Number of Detectors and Peak Tube Voltage? Radiology, September 1, 2008; 248(3): 910 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Israel and M. A. Bosniak Pitfalls in Renal Mass Evaluation and How to Avoid Them RadioGraphics, September 1, 2008; 28(5): 1325 - 1338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Birnbaum, N. Hindman, J. Lee, and J. S. Babb Renal Cyst Pseudoenhancement: Influence of Multidetector CT Reconstruction Algorithm and Scanner Type in Phantom Model Radiology, September 1, 2007; 244(3): 767 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Silverman, K. J. Mortele, K. Tuncali, M. Jinzaki, and E. S. Cibas Hyperattenuating Renal Masses: Etiologies, Pathogenesis, and Imaging Evaluation RadioGraphics, July 1, 2007; 27(4): 1131 - 1143. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Israel and M. A. Bosniak How I Do It: Evaluating Renal Masses Radiology, August 1, 2005; 236(2): 441 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Zalis, J. J. Perumpillichira, J. Y. Kim, C. Del Frate, C. Magee, and P. F. Hahn Polyp Size at CT Colonography after Electronic Subtraction Cleansing in an Anthropomorphic Colon Phantom Radiology, July 1, 2005; 236(1): 118 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Krakora, F. V. Coakley, G. Williams, B. M. Yeh, R. S. Breiman, and A. Qayyum Small Hypoattenuating Hepatic Lesions at Contrast-enhanced CT: Prognostic Importance in Patients with Breast Cancer Radiology, December 1, 2004; 233(3): 667 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Hecht, G. M. Israel, G. A. Krinsky, W. Y. Hahn, D. C. Kim, I. Belitskaya-Levy, and V. S. Lee Renal Masses: Quantitative Analysis of Enhancement with Signal Intensity Measurements versus Qualitative Analysis of Enhancement with Image Subtraction for Diagnosing Malignancy at MR Imaging Radiology, August 1, 2004; 232(2): 373 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Israel, N. Hindman, and M. A. Bosniak Evaluation of Cystic Renal Masses: Comparison of CT and MR Imaging by Using the Bosniak Classification System Radiology, May 1, 2004; 231(2): 365 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Suh, F. V. Coakley, A. Qayyum, B. M. Yeh, R. S. Breiman, and Y. Lu Distinction of Renal Cell Carcinomas from High-Attenuation Renal Cysts at Portal Venous Phase Contrast-enhanced CT Radiology, August 1, 2003; 228(2): 330 - 334. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
