Accuracy of polyp measurements depends on CT values

Polyps are most accurately measured at threshold values of approximately -500 HU, according to new Korean research that examined measurements at virtual colonoscopy (VC or CT colonography [CTC]). The study also found that the sizes of colorectal polyps varied significantly depending on the CT values used to define their contours.

The multicenter effort, led by researchers from the Ulsan College of Medicine in Seoul, also found that automated polyp measurements taken from 3D endoluminal views of VC data in a colon phantom are quite accurate compared to manual measurements, with similar findings in a small human cohort.

This last finding is evidence that automated polyp measurements, critical for determining patient management at virtual colonoscopy, are accurate when consistently performed, according to the authors, who believe that 3D endoluminal measurements offer intuitive value in determining what to measure.

"Although the current consensus proposal recommends that the single greatest dimension of the polyp should be obtained regardless of the use of 2D or 3D images, 3D endoluminal measurement is likely to have the advantage of being more suited in finding the long axis of the polyp, as one may instinctively visualize the long axis of the polyp by using the endoluminal view," wrote Dr. Seong Ho Park and colleagues. "The standard orthogonal 2D views, on the other hand, may fail to include the long axis. Finding an optimized MPR plane for each and every polyp is cumbersome and impractical for clinical practice. Three-dimensional endoluminal measurement, therefore, should become a very effective tool for polyp size estimation with further improvement in measurement accuracy" (Radiology, January 2008, Vol. 246:1, pp. 157-167).

Also participating in the study was Dr. Eugene Choi from Weill Medical College of Cornell University in New York City. Additional collaborating institutions included Hallym University College of Medicine and Kangnam Sacred Heart Hospital, and Seoul National University College of Medicine and Seoul National University Hospital, all in Seoul.

"Of the various factors that affect the accuracy of polyp measurement on endoluminal images, the threshold value for endoluminal rendering is most likely a critical factor," the researchers explained. "There is a gradual transition of attenuation values over several voxels at the colonic lumen-wall interface from the attenuation of air to that of soft tissue."

"A lower threshold value will shift the reconstructed luminal surface toward the air side, with a resulting expansion of rendered structures. Polyp size is, therefore, expected to increase with the decrease in the surface-rendering threshold value," they stated. "Although this relationship between object size and threshold value is well-established, to our knowledge, in no formal study has the optimal threshold value for endoluminal visualization of CT colonographic images for accurate polyp measurement yet been investigated."

The study aimed to find the optimal surface-rendering threshold value for 3D VC images for accurate manual polyp measurement, using direct measurement of simulated polyps as the reference standard. Additionally, the authors evaluated the agreement between manual and automated 3D measurements.

From pig colons, the team created 86 simulated polyps (reference size, 3-15 mm) for measurement using a validated method. A second human study cohort consisted of 10 patients (nine men and one woman, mean age 63.8 years) with 14 polyps.

The polyps, seen at virtual colonoscopy and confirmed at conventional colonoscopy, included one serrated adenoma, seven tubular adenomas, one tubular adenoma with high-grade dysplasia, and five adenocarcinomas 5-20 mm in diameter. Inclusion criteria included being located in a clean, well-distended area of the colon, free of image artifacts, and with available pathologic diagnosis following endoscopic removal.

The study included automated and manual polyp measurements with endoluminal views that were surface-rendered at threshold values of -800, -700, -600, and -500 HU, performed by a single observer. Bland-Altman analysis was used to test the agreement between VC measurements and reference sizes, and the agreement between automated and manual measurements.

The manual 3D endoluminal polyp measurements were performed using smooth surface-rendering parameters on an Advantage workstation (AW4.2_06, GE Healthcare, Chalfont St. Giles, U.K.). The automated measurements were performed on a Syngo Colonography PEV system (Siemens Medical Solutions, Malvern, PA). This latter system derives its polyp edge information not from threshold values, but by computing the gradient along the polyp border and determining the midpoint of edge transition, they explained. Curvature patterns are analyzed to determine where the polyp is attached to the colon wall.

According to the results, in the polyps fashioned from pig colons, the mean difference between the observed and reference sizes was 0.86 mm (95% limits of agreement: -0.52 mm, 2.24 mm), 0.55 mm (95% limits of agreement: -0.75 mm, 1.85 mm), 0.20 mm (95% limits of agreement: -1.11 mm, 1.50 mm), and -0.08 mm (95% limits of agreement: -1.43 mm, 1.27 mm) for -800, -700, -600, and -500 HU, respectively, the authors wrote. The mean measurement difference was 0.9 mm (95% limits of agreement: -1.49 mm, 1.67 mm) for automated measurements.

Measurement differences were unrelated to polyp size, the group noted. The manual polyp size measurements at -500 HU (p = 0.277) and automated polyp size (p = 0.288) were not significantly different from the reference size.

In 10 human polyps (after excluding from analysis four lesions that were large, lobulated, or adjacent to a haustral fold), automated measurement showed accurate demarcation of lesion boundaries, and automated measurements were closest to manual measurements at -500 HU.

The results demonstrate that polyp size is substantially affected by the surface-rendering threshold value used for endoluminal viewing of VC data, Park and colleagues wrote.

The conclusion is not definitive, they cautioned, because of the "technical infeasibility" of determining the precise reference size of in vivo polyps and the small number of lesions measured. Nevertheless, "the fact that automated measurement of the 10 human polyps (with boundaries that were accurately determined by using the automated software) was very close to the manual measurement at -500 HU, as it was with the simulated polyps, may provide indirect evidence that the conclusion should also hold for human polyps," they wrote.

An additional limitation is the complex and proprietary nature of surface-rendering protocols in 3D visualization; therefore, to an extent, all measurement of surface-rendered data is vendor-dependent, they explained.

To achieve uniformity between different 3D polyp-viewing systems, standardized calibration tools, such as a standardized phantom, may be necessary to optimize the rendering parameters for reproducibility across various systems.

Differences in determining long-axis measurements and the lack of multiple measurements for each polyp may have also contributed to the variability of results, the researchers wrote.

Polyp size generally declined as the threshold value increased from -800 to -500 HU, a finding that corresponds to the expected patterns on the basis of surface-rendering principles, they wrote, noting that small differences in polyp size can make an important difference in clinical practice.

"In CT colonographic screening protocols where polyp size dictates clinical treatment, this overlooked source of measurement error caused by surface-rendering threshold values for 3D visualization can cause problems," Park and colleagues wrote. "The clinical consequence of this error is overt, especially in patients who have polyps with sizes that are on the borderline between size categories (i.e., ≤ 5 mm, 6-9 mm, and ≥ 10 mm); in these patients, a slight size disparity may lead to changes in polyp size categorization, as indicated in our data."

In addition, automated polyp measurement is reproducible and agrees closely with manual endoluminal measurement at the optimal threshold value of -500 HU for well-circumscribed smooth, rounded polyps. "These findings may require further confirmation in a study in a large number of human polyps," they wrote.

By Eric Barnes staff writer
February 21, 2008

Related Reading

Colon polyps accurately measured using automated and manual 3D techniques, January 5, 2007

Reading method, insufflation affect polyp measurements, November 4, 2005

Polyp measurements more accurate in 3D, September 6, 2005

Copyright © 2008

Page 1 of 654
Next Page