HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2363041193v1 236/3/867    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kim, Y. J. Articles by Choi, B. I. Search for Related Content PubMed PubMed Citation Articles by Kim, Y. J. Articles by Choi, B. I.

Small-Bowel Obstruction in a Phantom Model of ex Vivo Porcine Intestine: Comparison of PACS Stack and Tile Modes for CT Interpretation¹

¹ From the Department of Radiology, Seoul National University College of Medicine, Institute of Radiation Medicine at Seoul National University Medical Research Center, and Clinical Research Institute at Seoul National University Hospital, 28 Yongon-dong, Chongno-gu, Seoul, 110-744, Korea. From the 2004 RSNA Annual Meeting. Received July 7, 2004; revision requested September 14; revision received October 11; accepted November 15. Address correspondence to J.K.H. (e-mail: hanjk{at}radcom.snu.ac.kr)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
PURPOSE: To compare computed tomographic (CT) image interpretation with picture archiving and communication systems (PACS) stack and tile modes for speed and accuracy of transition zone localization in small-bowel obstruction by using ex vivo porcine specimens.

MATERIALS AND METHODS: Twenty-five small-bowel obstruction phantom models made of ex vivo porcine intestines from a slaughterhouse were imaged at CT. One was used for observer training, and 24 were used for experimentation. At 20-cm intervals throughout the intestines, metallic markers were placed in the mesenteries immediately adjacent to bowel. One obstruction was made in each intestine, midway between markers, by ligating intestine with a 3-0 silk suture to simulate mechanical small-bowel obstruction. The lumen proximal to the ligation site was distended with air and a soybean oil–iodized oil mixture until at least two-thirds of the proximal intestine exceeded 2.0 cm in transverse diameter. Dilated segments were 310–550 cm in length. Soybean oil and a mixture of soybean and iodized oil were used to simulate differences in attenuation among bowel wall, intraluminal fluid, and extraluminal abdominal fat. Four experienced abdominal radiologists independently determined the transition zone by using stack mode (cine viewing of stacked images) and, at least 2 weeks later, tile mode (side-by-side image display). Accuracy and degree of error in counting markers were evaluated, and speed of interpretation was recorded. Statistical analysis was performed with the McNemar and Wilcoxon signed rank tests.

RESULTS: For all observers, accuracy of transition zone localization tended to be better with stack mode (63%–83% [15–20 phantoms]) than with tile mode (50%–63% [12–15 phantoms]), but the differences were not significant. For each observer, mean counting error was lower in stack mode (range, 0.96–2.48) than in tile mode (range, 1.74–3.22), with significance for three observers (P < .01, P < .01, and P = .04). Interpretation was significantly faster with stack mode by a factor of two to three for all observers (P < .01).

CONCLUSION: Stack mode evaluation for identification of the transition zone in obstructed small bowel is faster than evaluation with tile mode. Accuracy is not significantly different between modes, although there is a tendency toward better results with stack mode.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
With the advent of picture archiving and communication systems (PACS), soft-copy image interpretation can replace film-based hard-copy image interpretation. PACS workstations can provide two kinds of image display presentation: tile mode (conventional side-by-side image display) and stack mode (cine viewing of stacked images). Although stack mode display has been believed to be more convenient and to have theoretic advantages (1), scientific verification of its superiority has been limited. A number of studies performed to compare soft-copy and hard-copy interpretation of computed tomographic (CT) scans have showed results in favor of soft-copy image interpretation with use of the stack mode (2–5). However, since these comparisons were performed by using tile mode display with hard-copy images and stack mode display with soft-copy images, it is not entirely clear whether the results should be attributed to the superiority of stack mode display or to the inherent advantages of soft-copy image interpretation, such as free control of the window width and level settings. To our knowledge, there has been only one experimental investigation, with a single phantom, in which stack and tile mode displays have been compared in the common environment of soft-copy image interpretation (6).

In daily practice, stack mode display is endorsed to be more optimal than tile mode display in the evaluation of small-bowel obstruction, for which complex dilated segments of small bowel need to be traced for identification of the transition zone (7). However, to our knowledge, there has been no study to compare the two different display modes with regard to the radiologist's diagnostic performance in this setting. Thus, the purpose of our study was to compare PACS stack and tile mode CT scan interpretation for the evaluation time and accuracy of localization of the transition zone in small-bowel obstruction by using ex vivo porcine specimens.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Specimens
Twenty-five porcine small intestines were used for this study. The specimens were already excised from adult pigs (80–120 kg) that had been sacrificed at a local slaughterhouse. Most of small-bowel mesentery had been removed except for 2–3 cm contiguous to the bowel. Small intestines ranged 7–8 m in length. At every 20 cm along the intestine, commercially available 13 x 6-mm metallic office staples (Peace Korea, Seoul, Korea) were placed in the mesenteries as close to the bowel as possible, parallel to the long axis of the intestine. One obstruction was made in each intestine, midway between the staples, by ligating the intestine with 3-0 silk suture to simulate a mechanical obstruction (Fig 1). The mean length of the small intestine proximal to the obstruction was 424 cm ± 67 (standard deviation), with a range of 310–550 cm.

View larger version (127K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Porcine small intestine. Mechanical obstruction (arrow) was made by means of ligation with a 3-0 silk suture. The proximal dilated segment (white arrowheads) and distal collapsed segment (black arrowheads) are seen.

Each of the 25 specimens was placed into a specially constructed plastic container simulating the abdominal cavity, and an attempt was made to protect the dilated intestine from any kinking or abrupt folding. The abdominal cavity phantom was cylindrical with elliptical cross sections and was 17.0 cm in anteroposterior diameter, 20.5 cm in transverse diameter, 20.0 cm in craniocaudal length, and 0.9 cm in wall thickness (Fig 2). The phantom had a large square inlet, 6.3 x 20 cm, for convenient placement of the dilated bowel, as well as for easy manipulation of the specimen within the phantom. After deployment of the small bowel into the phantom, 100% edible soybean oil of about –110 HU was gently poured into the phantom until the phantom's cavity was filled with oil. The inlet of the plastic phantom was tightly covered with a plastic lid with two 17-mm holes that were 16 cm apart in craniocaudal fashion. One end of the deployed intestine was pulled out through each hole of the lid so that observers could easily identify the proximal and distal ends on CT images (Fig 3). All procedures for constructing the specimens were performed together by two authors (Y.J.K., K.R.S.)

View larger version (165K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Abdominal cavity phantom made of transparent acrylic material. It was cylindrical, with elliptical cross-sections. A lid with two holes tightly covered the rectangular inlet.

View larger version (132K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Small-bowel obstruction phantom. The distal collapsed segment (black arrow) and proximal dilated segment (white arrow) were pulled out of the two holes in the lid so that the observer could easily identify the respective ends on CT images.

CT Imaging
All phantoms were imaged with a four–detector row helical CT scanner (MX 8000; Marconi Medical Systems, Cleveland, Ohio) with the following parameters: section thickness of 2.5 mm, pitch of 1.25, reconstruction interval of 2 mm, rotation time of 0.5 second, 120 kVp, 250 mAs, and 230-mm field of view. All images were reconstructed with a 512 x 512 matrix. All images were stored in Digital Imaging and Communications in Medicine, or DICOM, format.

Image Assessment
The digital data were sent to a PACS server (Radmax; MaroTech, Seoul, Korea) and were distributed to workstations (Radmax; MaroTech). All images were downloaded onto the local hard drive of a display workstation before interpretation. All CT images were independently analyzed by four board-certified abdominal radiologists (S.H.K., J.Y.J., S.K.A., and C.J.H.), whose level of experience varied (range, 2–5 years of experience). All observers were well acquainted with the use of PACS. Training sessions were held before the interpretation sessions to allow the observers to become familiar with the phantom. Immediate feedback was provided during these sessions. Of 25 constructed phantoms, one was used for training purposes, and the remaining 24 were used for actual experiments. The four observers were asked to independently determine the transition zone by (a) counting the number of markers in the small intestine proximal to the ligature and (b) labeling which one of the eight segments showed the transition zone. Slice number was available in a given data set so that observers were able to determine segment boundaries in the cranial-caudal axis. Observers were informed that one ligation was made in each small-bowel phantom. An interpretation was considered to be accurate if the difference between the evaluated number and the reference number of markers was two or fewer and if the evaluated segment of transition zone was identical to the reference standard. The time required to assess the CT images was measured with a stopwatch by each observer. Observers were aware that both the speed of interpretation and the accuracy of localization of the transition zone would be assessed.

A PACS workstation (Radmax; Marotech) with three 2048 x 1536-pixel 20.8-inch monochrome liquid crystal display monitors (ME315L; Totoku Electric, Tokyo, Japan) was used for both stack and tile mode evaluation of the phantoms. The reviewers could freely control window width and level settings. In the stack display evaluation, manual stack mode was used with a computer mouse. In both evaluation modes, each observer was allowed to choose the number of images displayed per screen, but the preferred number was usually nine or 12 for the tile mode and one or four for the stack mode. An interval of at least 2 weeks was enforced between the stack mode and tile mode reading sessions. The CT images were presented to observers in a different order between the two reading sessions to minimize recall bias. All four observers performed stack mode evaluation before tile mode evaluation.

Statistical Analysis
For each observer, accuracy was calculated for both tile mode and stack mode evaluation. The degree of error was also calculated by using the difference between the number of markers counted by the observer and the actual number known for each phantom. Differences in accuracy for localization of the transition zone were evaluated with the McNemar test, and differences in speed of interpretation and degree of error in counting markers were assessed with the Wilcoxon signed rank test. Statistical analysis was performed by using SPSS for Windows (version 10.0; SPSS, Chicago, Ill), and P < .05 was considered to indicate a statistically significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Among all four observers, accuracy of localization of the transition zone tended to be better with stack mode display (63%–83% [15–20 of 24 phantoms]) than with tile mode display (50%–63% [12–15 phantoms]), but these differences were not statistically significant (Fig 4, Table 1). In the determination of the numbers of markers in dilated loops, the mean error was smaller with stack mode (range, 0.96–2.48) than with tile mode (range, 1.74–3.22), with a statistically significant difference for all but one observer (Table 2).

View larger version (111K): [in this window] [in a new window] [Download PPT slide]   Figure 4. CT image of a small-bowel obstruction phantom. The transition zone (black arrow) is identified. Multiple markers (white arrows) could be seen just outside the bowel wall.

All four observers evaluated the phantoms significantly faster with stack mode than with tile mode (Table 3). The difference between the two modes ranged from 2.0 to 3.0 times the speed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

The use of stack viewing or cine paging has been implemented to aid radiologists in performing such tasks more easily (7). It has been believed that any tubular or cordlike structures, such as arteries, veins, ducts, tendons, and nerves, could be traced with greater facility by using the display mode than by using the conventional tile mode (6). In an experimental study by Mathie and Strickland (6), performed with one phantom consisting of 20 intertwining tubes, tracing of such one-way directional tubes was performed three to five times faster with stack mode than with tile mode, and two mistakes occurred in a total of 50 tracings with tile mode display, while no mistakes occurred with stack mode.

Our results, derived from 24 different phantoms containing obstructed porcine small bowel, are in agreement with those of Mathie and Strickland (6). Improvement in speed by two- to threefold in our study was smaller than that in the study by Mathie and Strickland. We postulate that this can be attributed to the difference in the complexity of the phantoms used in the two studies: Following the convoluted porcine small-bowel loops in our study required greater effort, including backward and forward scrolling even with stacked images, in contrast to the unidirectional tracing of the intertwined plastic tubes in the other study.

The accuracy obtained with the stack mode evaluation was somewhat higher than that with tile mode, but this difference was not statistically significant. Nonetheless, when counting the numbers of markers in dilated proximal bowel loops, the degree of error was significantly smaller with stack mode than with tile mode for all but one observer (Table 2). Hence, these findings, taken in the context of improved speed with stack mode interpretation, suggest the superiority of stack mode evaluation in tracing tubular structures such as bowels. Our results agree with those of other clinical studies involving stack mode, for which a certain degree of superiority of stack mode display has been reported (although not always with statistical significance) (2–5). However, it should be noted that there are two notable differences between our study and previous clinical studies: (a) authors of such clinical studies did not compare the two modes in a common soft-copy image environment, and (b) those studies involved the assessment of only nodular structures, such as lung nodules and liver metastases.

The improved speed with stack mode interpretation can be attributed to several factors (1,6). First, stack mode navigation reveals longitudinal relationships between contiguous images more clearly so that the observer can trace tubular structures coursing perpendicular to the transverse plane with ease and confidence. Second, this mode considerably reduces the observer's eye and head movement, which is required for tile mode interpretation, especially with large data sets of images. With the stack mode, radiologists can maintain their focus on a specific spatial location as images are scrolled, while tile mode requires observers to constantly shift their gaze between multiple images displayed in separated spatial fields. Since evaluation of a cross-sectional imaging study requires mental integration of multiple sections into three-dimensional information, stack mode is more natural and intuitive than is the conventional tile mode, in which such information becomes fragmented by being displayed side by side. Results of a recent study showed that radiologists are using the stack mode more frequently in their CT scan interpretation compared with the beginning era of PACS implementation (11).

In our study, we allowed the observers to freely set the number of images per screen so that larger images could be used with the stack mode. Indeed, allowance of larger image sizes is one of inherent advantages of stack mode interpretation. A related merit of stack mode display is that fewer monitors are required than for tile mode display.

There were several limitations in our phantom study. First, we used a whole porcine small bowel that was 700–800 cm in length, and we made dilated segments ranging from 310 to 550 cm. Considering that the normal human small bowel is 350–700 cm in length (12), our phantom might have required the observers to trace bowel segments longer than is required for actual cases in humans. On the other hand, our phantom did not have any colon, visceral organs, or mesenteric vessels, which can often complicate identification of the transition zone in actual clinical practice. Given these two opposing factors with regard to difficulty of interpretation, we believe that the overall level of difficulty in our phantom is likely similar to that experienced in actual clinical settings.

Second, markers were placed at the outside of the small bowel wall—albeit immediately adjacent to the bowel wall—so that it was sometimes difficult to distinguish which marker represented which segment of bowel when they were seen between two adjacent, abutting segments. This difficulty could lead to error in counting the number of markers, thus biasing the accuracy toward underestimation. The observers were informed that the staple markers were positioned parallel to the long axis of the bowel, so that they could resolve this difficulty when two abutting bowel segments were coursing at different angles. Because radiologists usually pay much more attention to the center point of tubular structures than to the periphery when following dilated bowel, this may also compound any error in determining the numbers of markers in dilated segments. In determining accuracy, we used two kinds of references, which consisted of the number of markers, with some degree of flexibility (–2 to +2), and the segment where the transition zone is seen; therefore, we think that such counting errors were minimized.

Third, since we made a complete ligation in every phantom, our experimental study could not be used to appropriately model partial obstruction or paralytic ileus. In addition, given the lack of negative controls, this study could not be used to assess the entirety of the radiologist's decision-making process for complete bowel obstruction.

Finally, there was a potential for reader bias, because the observers themselves were asked to measure the time taken for interpretation (although a stopwatch was used). Also, since all observers performed the stack mode evaluation first, they might have been more adjusted to the phantoms at the time of tile mode evaluation even if they did not recall the result of the previous evaluation.

In summary, we compared radiologists' performance in identifying the transition zone in small-bowel obstruction between two different display modes in PACS. Our results clearly indicate that stack mode evaluation is faster than tile mode evaluation. The accuracy with stack mode also tended to be better than that with tile mode; however, this difference did not reach criteria for statistical significance.

Practical application: Stack mode is preferred for CT image evaluation of small-bowel obstruction because it can facilitate localization of the transition zone, which is usually the most time-consuming and challenging step in the diagnosis of small-bowel obstruction at CT. Also, cross-sectional imaging interpretation of other structures with three-dimensional spatial complexity may benefit from the stack mode display.

	ACKNOWLEDGMENTS

 
We thank Nam C. Yu, MD (Department of Radiology, University of California, Los Angeles), for the linguistic revision and for general discussion, and Seong Ho Park, MD (Department of Radiology, Asan Medical Center, Seoul, Korea), for his invaluable statistical consultation.

	FOOTNOTES

 

Abbreviations: PACS = picture archiving and communication system

² Current address: Department of Radiology, Konkuk University School of Medicine, Seoul, Korea

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, J.K.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, Y.J.K.; experimental studies, Y.J.K., K.R.S.; statistical analysis, J.M.L.; and manuscript editing, Y.J.K., J.K.H., K.H.L., J.M.L.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Reiner BI, Siegel EL, Siddiqui K. Evolution of the digital revolution: a radiologist perspective. J Digit Imaging 2003;16:324–330.[CrossRef][Medline]
2. Seltzer SE, Judy PF, Adams DF, et al. Spiral CT of the chest: comparison of cine and film-based viewing. Radiology 1995;197:73–78.[Abstract/Free Full Text]
3. Tillich M, Kammerhuber F, Reittner P, Riepl T, Stoeffler G, Szolar DH. Detection of pulmonary nodules with helical CT: comparison of cine and film-based viewing. AJR Am J Roentgenol 1997;169:1611–1614.[Abstract/Free Full Text]
4. Iwano S, Makino N, Ikeda M, et al. Videotaped helical CT images for lung cancer screening. J Comput Assist Tomogr 2000;24:242–246.[CrossRef][Medline]
5. Pijl ME, Wasser MN, Joekes EC, van de Velde CJ, Bloem JL. Metastases of colorectal carcinoma: comparison of soft- and hard-copy helical CT interpretation. Radiology 2003;227:747–751.[Abstract/Free Full Text]
6. Mathie AG, Strickland NH. Interpretation of CT scans with PACS image display in stack mode. Radiology 1997;203:207–209.[Abstract/Free Full Text]
7. Memel DS, Berland LL. CT of bowel obstruction: interpretation using cine-paging. AJR Am J Roentgenol 1995;164:766–767.[Medline]
8. Furukawa A, Yamasaki M, Furuichi K, et al. Helical CT in the diagnosis of small bowel obstruction. RadioGraphics 2001;21:341–355.[Abstract/Free Full Text]
9. Khurana B, Ledbetter S, McTavish J, Wiesner W, Ros PR. Bowel obstruction revealed by multidetector CT. AJR Am J Roentgenol 2002;178:1139–1144.[Free Full Text]
10. Gazelle GS, Goldberg MA, Wittenberg J, Halpern EF, Pinkney L, Mueller PR. Efficacy of CT in distinguishing small-bowel obstruction from other causes of small-bowel dilatation. AJR Am J Roentgenol 1994;162:43–47.[Abstract/Free Full Text]
11. Bennett WF, Vaswani KK, Mendiola JA, Spigos DG. PACS monitors: an evolution of radiologist's viewing techniques. J Digit Imaging 2002; 15(suppl 1):171–174.
12. Weaver LT, Austin S, Cole TJ. Small intestinal length: a factor essential for gut adaptation. Gut 1991;32:1321–1323.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. H. Kim, J. M. Lee, Y. J. Kim, J. Y. Choi, G. H. Kim, H. Y. Lee, and B. I. Choi Detection of Hepatocellular Carcinoma on CT in Liver Transplant Candidates: Comparison of PACS Tile and Multisynchronized Stack Modes Am. J. Roentgenol., May 1, 2007; 188(5): 1337 - 1342. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2363041193v1 236/3/867    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kim, Y. J. Articles by Choi, B. I. Search for Related Content PubMed PubMed Citation Articles by Kim, Y. J. Articles by Choi, B. I.