(Radiology. 2000;214:517-522.)
© RSNA, 2000
Mucosal Detail at CT Virtual Reality: Surface versus Volume Rendering1
Kenneth D. Hopper, MD,
A. Tunç Iyriboz, MD,
Scott W. Wise, MD,
Joel D. Neuman, MD,
David T. Mauger, PhD and
Claudia J. Kasales, MD
1 From the Departments of Radiology (K.D.H., A.T.I., S.W.W., C.J.K.) and Health Evaluation Sciences (D.T.M.), Milton S. Hershey Medical Center, Pennsylvania State University, 500 University Dr, Hershey, PA 17033; and the Department of Radiology, Geisinger Wyoming Valley Hospital, Wilkes Barre, Pa (J.D.N.). Received January 7, 1998; revision requested April 2; revision received April 13, 1999; accepted June 10. Address reprint requests to K.D.H. (e-mail: khopper@psu.edu).
 |
Abstract
|
|---|
PURPOSE: To evaluate computed tomographic virtual reality with volumetric versus surface rendering.
MATERIALS AND METHODS: Virtual reality images were reconstructed for 27 normal or pathologic colonic, gastric, or bronchial structures in four ways: the transition zone (a) reconstructed separately from the wall by using volume rendering; (b) with attenuation equal to air; (c) with attenuation equal to wall (soft tissue); (d) with attenuation halfway between air and wall. The four reconstructed images were randomized. Four experienced imagers blinded to the reconstruction graded them from best to worst with predetermined criteria.
RESULTS: All readers rated images with the transition zone as a separate structure as overwhelmingly superior (P < .001): Nineteen cases had complete concurrence among all readers. The best of the surface-rendering reconstructions had the transition zone attenuation equal to the wall attenuation (P < .001). The third best reconstruction had the transition zone attenuation equal to the air attenuation, and the worst had the transition zone attenuation halfway between the air and wall attenuation.
CONCLUSION: Virtual reality is best with volume rendering, with the transition zone (mucosa) between the wall and air reconstructed as a separate structure.
Index terms: Bronchi, CT, 671.12117 • Colon, CT, 75.12117 • Computed tomography (CT), image quality, 671.12117, 72.12117, 75.12117 • Computed tomography (CT), three-dimensional, 671.12117, 72.12117, 75.12117 • Computed tomography (CT), volume rendering, 671.12117, 72.12117, 75.12117 • Images, processing, 671.12117, 72.12117, 75.12117 • Stomach, CT, 72.12117 • Trachea, CT, 671.12117
|
Introduction
|
|---|
Computed tomographic (CT) virtual reality targets the air-mucosal interface in the creation of endoscopic, lifelike images. Frequently, however, CT virtual reality has been performed with surface rendering under the assumption that the opacity change between the wall and lumen (air) is instantaneous. The transition between these widely separated attenuation coefficients, however, actually occurs over several voxels.
Using volumetric techniques, we have learned to reconstruct this transition zone (mucosa) separately from the wall and lumen, which allows a dramatic improvement in the delineation of mucosal detail. In this project, we compared CT virtual reality mucosal detail obtained with volume rendering versus three different surface-rendering techniques by using multiple readers blinded to the reconstruction technique.
|
MATERIALS AND METHODS
|
|---|
All CT scanning was performed with a 2–4-mm section thickness, 50% overlap of the reconstructed sections, 175–225 mA, 120 kV, and a standard interpolator and algorithm (scanner model PQ5000; Picker International, Cleveland, Ohio). For the normal stomach cases, five air-distended pig stomachs were used because of their prominent folds. Five normal sheep trachea and bronchi were used because of the availability of gross mounted specimens and their similarity to the human airway. A preponderance (n = 17) of human colonic cases (normal or abnormal) were included, given the current emphasis on CT virtual reality of the colon. The specific sites and abnormalities are listed in Table 1.
The CT data were loaded onto a workstation (Voxel Q; Picker International). Between the luminal wall and the intraluminal air, there was a third layer of transition voxels with attenuation coefficients between the attenuation of the wall and that of the air. For volume rendering, these transition voxels, thought to largely represent the mucosa, were reconstructedseparately from the wall with a 50% opacity. To define the displayed surface, for the three surface-rendering techniques, three attenuation coefficients that included with the colonic wall all, half, or none of the transition voxels were chosen (Fig 1).
View larger version (21K):
[in this window]
[in a new window]
|
Figure 1. Diagram of the four reconstruction techniques. Method 1 is volume rendering, with the mucosa (TZ) reconstructed separately from the wall. Methods 2-4 are surface rendering with the mucosal (TZ) attenuation either made equal to the air attenuation (method 2) or the wall attenuation (method 3) or made to be halfway between the two (method 4).
|
|
In reconstruction method 1, the volume-rendering method, the transition zone (mucosa) was reconstructed as a separate structure (Fig 2). In method 2, a surface-rendering method, the attenuation of the transition zone (mucosa) was set to be equal to the attenuation of air. In method 3, a surface-rendering method, the attenuation of the transition zone (mucosa) was set to be equal to the attenuation of the wall (soft tissue). In method 4, a surface-rendering method, the attenuation of the transition zone (mucosa) was set to be halfway between the attenuation of air and that of the wall (soft tissue).
View larger version (176K):
[in this window]
[in a new window]
|
Figure 2a. When reconstructed by itself by using volumetric rendering, the transition zone (mucosa) appears as a translucent membrane with a thickness of 1-3 voxels. (a) Loop of colon. (b) View of the carina.
|
|
View larger version (162K):
[in this window]
[in a new window]
|
Figure 2b. When reconstructed by itself by using volumetric rendering, the transition zone (mucosa) appears as a translucent membrane with a thickness of 1-3 voxels. (a) Loop of colon. (b) View of the carina.
|
|
Each of the four reconstructed images of each case was assigned a computer-generated random number. Within each set, the four images were arranged in ascending numerical order. The image sets then were submitted, along with photographs of the mounted gross colon, pig stomach, and sheep airway, to five experienced imagers (K.D.H., A.T.I., S.W.W., J.D.N., C.J.K.) blinded to the reconstruction technique. Working independently, each reader used the gross specimen photographs plus the reader's medical experience to rank the four virtual images in each set from the best (most realistic) to the worst on a scale of 1 to 4, with 1 denoting best.
In addition, the readers used lesion conspicuity when a lesion was present, image sharpness, amount of near- and far-field blurring, and degree of stair-step artifact to evaluate the images and rank them from best to worst, with 1 denoting best. The rankings for the 27 image sets were entered onto a data sheet by each reader and were submitted for biostatistical analysis (total measurements: 27 sets x 4 readers = 108). After each reading, the images in each set were rearranged in ascending numerical order. This was verified by a research assistant prior to submission to another reader.
The Fisher exact test was performed to assess differences in the relative ranking of the four reconstruction methods. Reader and tissue effects were assessed by using stratification.
|
RESULTS
|
|---|
The images on which the transition zone (mucosa) was reconstructed as a separatestructure by using volume rendering (reconstruction method 1) overwhelmingly were rated as superior by all four readers (Fig 3–6). Overall, method 1 was ranked best in 98 of 108 (27 structures x 4 readers) comparisons (91%). In addition, in 19 of the 27 cases, there was complete concurrence among all four readers that this reconstruction method was the best. Reconstruction of the transition zone (mucosa) as a separate structure by using volume rendering was significantly superior to the three surface-rendering reconstruction methods (P < .001).
View larger version (148K):
[in this window]
[in a new window]
|
Figure 3a. Case 15. Reconstructions in a patient with two polyps (arrows in a) smaller than 5 mm in the descending colon that were verified at colonoscopy (colonoscopic images not shown). All four readers rated (a) reconstruction method 1 (transition zone, or mucosa, reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.25, 2.25, and 3.50, respectively. (e) Color rendition of the polyps with reconstruction method 1 (a) shows the polyps in excellent relief.
|
|
View larger version (195K):
[in this window]
[in a new window]
|
Figure 3b. Case 15. Reconstructions in a patient with two polyps (arrows in a) smaller than 5 mm in the descending colon that were verified at colonoscopy (colonoscopic images not shown). All four readers rated (a) reconstruction method 1 (transition zone, or mucosa, reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.25, 2.25, and 3.50, respectively. (e) Color rendition of the polyps with reconstruction method 1 (a) shows the polyps in excellent relief.
|
|
View larger version (195K):
[in this window]
[in a new window]
|
Figure 3c. Case 15. Reconstructions in a patient with two polyps (arrows in a) smaller than 5 mm in the descending colon that were verified at colonoscopy (colonoscopic images not shown). All four readers rated (a) reconstruction method 1 (transition zone, or mucosa, reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.25, 2.25, and 3.50, respectively. (e) Color rendition of the polyps with reconstruction method 1 (a) shows the polyps in excellent relief.
|
|
View larger version (193K):
[in this window]
[in a new window]
|
Figure 3d. Case 15. Reconstructions in a patient with two polyps (arrows in a) smaller than 5 mm in the descending colon that were verified at colonoscopy (colonoscopic images not shown). All four readers rated (a) reconstruction method 1 (transition zone, or mucosa, reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.25, 2.25, and 3.50, respectively. (e) Color rendition of the polyps with reconstruction method 1 (a) shows the polyps in excellent relief.
|
|
View larger version (147K):
[in this window]
[in a new window]
|
Figure 3e. Case 15. Reconstructions in a patient with two polyps (arrows in a) smaller than 5 mm in the descending colon that were verified at colonoscopy (colonoscopic images not shown). All four readers rated (a) reconstruction method 1 (transition zone, or mucosa, reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.25, 2.25, and 3.50, respectively. (e) Color rendition of the polyps with reconstruction method 1 (a) shows the polyps in excellent relief.
|
|
A pattern emerged as to the relative ranking of the three surface-rendering reconstruction methods (Tables 1, 2) For the airway, stomach, colon, and the combined data, the third reconstruction method (in which the transition zone attenuation equaled the wall attenuation) proved superior to the other two (P < .001). The third best reconstruction method was that in which the transition zone (mucosal) attenuation was set equal to the air attenuation (method 2), and the worst was that in which the transition zone (mucosal) attenuation was halfway between air attenuation and wall attenuation (method 4). This latter finding, however, was true only in the airway, colon, and in the combined data. For the stomach, the worst reconstruction method was that in which the transition zone (mucosal) attenuation was set equal to air attenuation. The overall mean score for volume rendering with the transition zone (mucosa) reconstructed as a separate structure was 1.19, versus 3.04, 2.50, and 3.27 for surface-rendering reconstruction methods 2–4.
|
DISCUSSION
|
|---|
Human perception is directly related to the informational content of the stimulus (1). However, small stimuli are perceived 25% less often per cubic milliliter of volume than large stimuli (2). The perception of small stimuli is related not only to the background and visual luminance but also to higher-level mental abstracting by the reviewer (3). Factors that delineate small foci from their background and increase their conspicuity (stimulus) to the reader include sharpness, shadow, depth, and color (1–4).
To our knowledge, there are two rendering techniques currently used to produce CT virtual reality images. In thresholding, or surface rendering, a specific attenuation coefficient is used to define the air-lumen interface. Simple to use and able to be performed on inexpensive computer platforms, surface rendering then assigns all structures either into air or wall, depending on the attenuation coefficient selected. With a slightly higher threshold, the transition zone voxels will be assigned into the lumen (air). With a slightly lower threshold, these will be assigned into the wall. Because of its nature, the transition zone cannot be targeted as a separate structure with surface rendering. In addition, the optimal single attenuation coefficient that differentiates wall from lumen varies between patients and pathologic findings, which further decreases the detail seen on the image. Last, surface rendering is sensitive to artifact and noise. As a result, surface rendering and its ubiquitous volume averaging (5), variability, and absence or exaggeration of detail make it undesirable for use in CT virtual reality.
With volume rendering, the entire spiral CT volume is classified, which allows voxels to be grouped into multiple categories on the basis of their attenuation coefficients and the groups to be reconstructed as separate structures. Unlike surface rendering, volume rendering allows the transition voxels between the air and the wall to be reconstructed separately as a specific structure. While requiring far more computer power, volume rendering with the delineation of the transition zone as a separate structure significantly enhances mucosal (and lesion) detail, adds depth and three-dimensional relief to the images, and allows the inclusion of lifelike complex colors with shading and shadows (5–8). The volume-rendering images used in this study can be created on any workstation with volume-rendering software that allows at least three CT ranges (air, wall, transition zone) to be reconstructed simultaneously and that allows the opacity of each to be varied.
Spiral CT offers unique advantages for virtual reality imaging. A characteristic of spiral CT is its improved resolution of the z axis (into the gantry and parallel to the long axis of the patient) (9,10). This is very important in any three-dimensional CT application because the section thickness is several times larger than the pixel size. Spiral CT also eliminates section-to-section (breath-to-breath) misregistration. Last, spiral CT allows an unlimited overlapping of reconstructed sections with no additional radiation, unlike conventional CT, with consecutive, nonoverlapped sections.
The results of this study strongly support volume rendering for CT virtual reality, with the transition zone (mucosa) reconstructed as a separate structure. CT virtual reality can be used to achieve excellent mucosal and lesion detail when volume rendering is used, with the transition zone (mucosa) reconstructed separately from the wall and the lumen. In addition, the reconstruction with volume rendering of the transition zone (mucosa) as a separate structure allows the use of lifelike colors, which adds image depth and lesion relief to the virtual images. The conspicuity (stimulus) of small lesions and our perception of them is, as a result, improved.
View larger version (140K):
[in this window]
[in a new window]
|
Figure 4a. Case 3. Reconstructions obtained in a patient with a substantial amount of stool in the transverse colon. All four readers rated (a) reconstruction method 1 (transition zone reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 2.75, 3.50, and 2.75, respectively.
|
|
View larger version (166K):
[in this window]
[in a new window]
|
Figure 4b. Case 3. Reconstructions obtained in a patient with a substantial amount of stool in the transverse colon. All four readers rated (a) reconstruction method 1 (transition zone reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 2.75, 3.50, and 2.75, respectively.
|
|
View larger version (164K):
[in this window]
[in a new window]
|
Figure 4c. Case 3. Reconstructions obtained in a patient with a substantial amount of stool in the transverse colon. All four readers rated (a) reconstruction method 1 (transition zone reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 2.75, 3.50, and 2.75, respectively.
|
|
View larger version (165K):
[in this window]
[in a new window]
|
Figure 4d. Case 3. Reconstructions obtained in a patient with a substantial amount of stool in the transverse colon. All four readers rated (a) reconstruction method 1 (transition zone reconstructed separately) as the best, with a mean score of 1.00. (b-d) Reconstruction methods (b) 2, (c) 3, and (d) 4 had mean scores of 2.75, 3.50, and 2.75, respectively.
|
|
View larger version (157K):
[in this window]
[in a new window]
|
Figure 5a. Case 20. Reconstructions in a normal pig stomach. All four readers agreed that (a) reconstruction method 1 was the best, with a mean score of 1.00. (b-d) Mean scores for methods (b) 2, (c) 3, and (d) 4 were 3.00, 2.75, and 3.25, respectively.
|
|
View larger version (182K):
[in this window]
[in a new window]
|
Figure 5b. Case 20. Reconstructions in a normal pig stomach. All four readers agreed that (a) reconstruction method 1 was the best, with a mean score of 1.00. (b-d) Mean scores for methods (b) 2, (c) 3, and (d) 4 were 3.00, 2.75, and 3.25, respectively.
|
|
View larger version (184K):
[in this window]
[in a new window]
|
Figure 5c. Case 20. Reconstructions in a normal pig stomach. All four readers agreed that (a) reconstruction method 1 was the best, with a mean score of 1.00. (b-d) Mean scores for methods (b) 2, (c) 3, and (d) 4 were 3.00, 2.75, and 3.25, respectively.
|
|
View larger version (177K):
[in this window]
[in a new window]
|
Figure 5d. Case 20. Reconstructions in a normal pig stomach. All four readers agreed that (a) reconstruction method 1 was the best, with a mean score of 1.00. (b-d) Mean scores for methods (b) 2, (c) 3, and (d) 4 were 3.00, 2.75, and 3.25, respectively.
|
|
View larger version (172K):
[in this window]
[in a new window]
|
Figure 6a. Case 25. Reconstructions in normal sheep bronchi oriented so that the viewer is looking down the right main bronchus. The middle lobe bronchus is on the viewer's left and the large lower lobe bronchus is on the viewer's right. All four readers ranked (a) reconstruction method 1 as the best, with a mean score of 1.00. (b-d) Methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.00, 2.00, and 4.00, respectively.
|
|
View larger version (203K):
[in this window]
[in a new window]
|
Figure 6b. Case 25. Reconstructions in normal sheep bronchi oriented so that the viewer is looking down the right main bronchus. The middle lobe bronchus is on the viewer's left and the large lower lobe bronchus is on the viewer's right. All four readers ranked (a) reconstruction method 1 as the best, with a mean score of 1.00. (b-d) Methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.00, 2.00, and 4.00, respectively.
|
|
View larger version (203K):
[in this window]
[in a new window]
|
Figure 6c. Case 25. Reconstructions in normal sheep bronchi oriented so that the viewer is looking down the right main bronchus. The middle lobe bronchus is on the viewer's left and the large lower lobe bronchus is on the viewer's right. All four readers ranked (a) reconstruction method 1 as the best, with a mean score of 1.00. (b-d) Methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.00, 2.00, and 4.00, respectively.
|
|
View larger version (200K):
[in this window]
[in a new window]
|
Figure 6d. Case 25. Reconstructions in normal sheep bronchi oriented so that the viewer is looking down the right main bronchus. The middle lobe bronchus is on the viewer's left and the large lower lobe bronchus is on the viewer's right. All four readers ranked (a) reconstruction method 1 as the best, with a mean score of 1.00. (b-d) Methods (b) 2, (c) 3, and (d) 4 had mean scores of 3.00, 2.00, and 4.00, respectively.
|
|
|
Footnotes
|
|---|
Author contributions: Guarantor of integrity of entire study, K.D.H.; study concepts and design, K.D.H.; definition of intellectual content, K.D.H.; literature research, K.D.H.; clinical and experimental studies, K.D.H., A.T.I., S.W.W., J.D.N, C.J.K.; data acquisition, K.D.H., A.T.I., S.W.W., J.D.N, C.J.K.; data analysis, D.T.M.; statistical analysis, D.T.M.; manuscript preparation, K.D.H., D.T.M.; manuscript editing and review, all authors.
|
References
|
|---|
-
Norwich KH. On the theory of Weber fractions. Percept Psychophys 1987; 42:286-298.[Medline]
-
Ekman G, Junge K. Psychophysical relations in visual perception of length, area and volume. Scand J Psychol 1961; 2:1-10.
-
Burbeck CA, Hadden S. Scaled position integration areas: accounting for Weber's law for separation. Opt Soc Am A 1993; 10:5-15.
-
Seltzer SE, Cavanagh P, Judy PF, Swensson RG, Scarff L, Monsky W. Enhanced displays of medical images: evaluation of the effectiveness of color, motion, and contour for detecting and localizing liver lesions. Acad Radiol 1995; 2:748-755.[Medline]
-
Fishman EK, Magid D, Ney DR, et al. Three-dimensional imaging. Radiology 1991; 181:321-337.[Free Full Text]
-
Fishman EK, Ney DR. Advanced computer applications in radiology: clinical applications. RadioGraphics 1993; 13:463-475.[Abstract]
-
Drebin RA, Carpenter L, Hanrahan P. Volume rendering. Comput Graphics 1988; 22:65-74.
-
Levoy M. Display of surfaces from volume data. IEEE Comput Graph Appl 1988; 8:29-37.
-
Kalender WA, Polacin A, Süss C. A comparison of conventional and spiral CT: an experimental study on the detection of spherical lesions. J Comput Assist Tomogr 1994; 18:167-176.[Medline]
-
Kasales CJ, Hopper KD, Ariola DN, et al. Reconstructed helical CT scan: improvement in z-axis resolution compared with overlapped and nonoverlapped conventional CT scans. AJR Am J Roentgenol 1995; 164:1281-1284.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
S. H. Park, S. S. Lee, J. K. Kim, M.-J. Kim, H. J. Kim, S. Y. Kim, M.-Y. Kim, A. Y. Kim, and H. K. Ha
Volume Rendering with Color Coding of Tagged Stool during Endoluminal Fly-through CT Colonography: Effect on Reading Efficiency
Radiology,
September 1, 2008;
248(3):
1018 - 1027.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Tsampoulas, A. C. Tsili, D. Giannakis, Y. Alamanos, N. Sofikitis, and S. C. Efremidis
16-MDCT Cystoscopy in the Evaluation of Neoplasms of the Urinary Bladder
Am. J. Roentgenol.,
March 1, 2008;
190(3):
729 - 735.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. H. Park, E. K. Choi, S. S. Lee, J. Y. Woo, S. Y. Chung, Y. J. Kim, J. K. Han, and H. K. Ha
Linear Polyp Measurement at CT Colonography: 3D Endoluminal Measurement with Optimized Surface-rendering Threshold Value and Automated Measurement
Radiology,
December 1, 2007;
246(1):
157 - 167.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Burling, S. Halligan, S. Taylor, D. J. Brennand, D. G. Altman, P. Bassett, W. Atkin, and C. I. Bartram
Polyp measurement using CT colonography: agreement with colonoscopy and effect of viewing conditions on interobserver and intraobserver agreement.
Am. J. Roentgenol.,
June 1, 2006;
186(6):
1597 - 1604.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Shitrit, P. Valdsislav, A. Grubstein, D. Bendayan, M. Cohen, and M. R. Kramer
Accuracy of Virtual Bronchoscopy for Grading Tracheobronchial Stenosis: Correlation With Pulmonary Function Test and Fiberoptic Bronchoscopy
Chest,
November 1, 2005;
128(5):
3545 - 3550.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y. Onodera, T. Omatsu, S. Takeuchi, N. Shinagawa, K. Yamazaki, T. Nishioka, and K. Miyasaka
Enhanced Virtual Bronchoscopy Using the Pulmonary Artery: Improvement in Route Mapping for Ultraselective Transbronchial Lung Biopsy
Am. J. Roentgenol.,
October 1, 2004;
183(4):
1103 - 1110.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. De Wever, V. Vandecaveye, S. Lanciotti, and J.A. Verschakelen
Multidetector CT-generated virtual bronchoscopy: an illustrated review of the potential clinical indications
Eur. Respir. J.,
May 1, 2004;
23(5):
776 - 782.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
X. Chen, Z. Ai, M. Rasmussen, P. Bajcsy, L. Auvil, M. Welge, L. Leach, S. Vangveeravong, A. J. Maniotis, and R. Folberg
Three-Dimensional Reconstruction of Extravascular Matrix Patterns and Blood Vessels in Human Uveal Melanoma Tissue: Techniques and Preliminary Findings
Invest. Ophthalmol. Vis. Sci.,
July 1, 2003;
44(7):
2834 - 2840.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. M. Horton and E. K. Fishman
Current Role of CT in Imaging of the Stomach
RadioGraphics,
January 1, 2003;
23(1):
75 - 87.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. K. Kim, J. H. Ahn, T. Park, H. J. Ahn, C. S. Kim, and K.-S. Cho
Virtual Cystoscopy of the Contrast Material--Filled Bladder in Patients with Gross Hematuria
Am. J. Roentgenol.,
September 1, 2002;
179(3):
763 - 768.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. D. Cody
AAPM/RSNA Physics Tutorial for Residents: Topics in CT: Image Processing in CT
RadioGraphics,
September 1, 2002;
22(5):
1255 - 1268.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. M. Boiselle, K. F. Reynolds, and A. Ernst
Multiplanar and Three-Dimensional Imaging of the Central Airways with Multidetector CT
Am. J. Roentgenol.,
August 1, 2002;
179(2):
301 - 308.
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. M. Boiselle and A. Ernst
Recent Advances in Central Airway Imaging*
Chest,
May 1, 2002;
121(5):
1651 - 1660.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A S Chaoui, M A Blake, and T Glucker
CT colonography and colorectal screening
Gut,
February 1, 2002;
50(2):
284 - 284.
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. T. Ferrucci
Colon Cancer Screening with Virtual Colonoscopy: Promise, Polyps, Politics
Am. J. Roentgenol.,
November 1, 2001;
177(5):
975 - 988.
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. M. Horton and E. K. Fishman
Multi-Detector Row CT of Mesenteric Ischemia: Can It Be Done?
RadioGraphics,
November 1, 2001;
21(6):
1463 - 1473.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. D. Hopper, T. A. Lucas, K. Gleeson, J. L. Stauffer, R. Bascom, D. T. Mauger, and R. Mahraj
Transbronchial Biopsy with Virtual CT Bronchoscopy and Nodal Highlighting
Radiology,
November 1, 2001;
221(2):
531 - 536.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
