| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Michigan, 1500 E Medical Center Dr.,
Ann Arbor, MI 48109.
2 Department of Biostatistics, School of Public Health, University of Michigan,
Ann Arbor, MI.
3 Division of Hematology/Oncology, Department of Internal Medicine, University
of Michigan, Ann Arbor, MI.
Received June 21, 2005;
accepted after revision February 14, 2006.
Address correspondence to M. Petrou
(mpetrou{at}umich.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Seventy-five individual nodules identified on 29 lowdose, unenhanced, MDCT chest examinations were selected for volumetric analysis. Each image data set was reconstructed in three ways (slice thickness/reconstruction interval): 1.25 mm/0.625 mm, 2.5 mm/2 mm, and 5 mm/2.5 mm. Volumetric measurements were made on all 75 nodules at 1.25- and 2.5-mm slice thicknesses and on 57 of 75 nodules at the 5-mm thickness using Volume Analysis software. For 69 of 75 nodules, measurements were obtained on 1.25- and 2.5-mm-thick sections using a different commercially available software system, LN500 R2 software. Volume variability between different slice thicknesses was correlated with nodule diameter, shape, and margination using multiple linear regression. Percent differences between measurements obtained with the two software systems were calculated. Significance of relative volume differences between slice thicknesses and software packages was assessed using a one-sample Student's t-test.
RESULTS. Although statistically significant differences in volumes between different section thicknesses were seen only for the tiny nodule size group, many individual nodules showed substantial volume variation. Significant differences were seen in nodule volume variability for smaller nodules (3-10 mm) compared with larger nodules (3 11 mm) (p < 0.0001), as well as spiculated compared with smooth nodules, within a single size group (p < 0.05). No effect of nodule shape (round vs elongated) was noted. Statistically significant differences in measurements obtained with the two software systems were seen only with 2.5-mm-thick sections (p = 0.001).
CONCLUSION. CT slice thickness variation resulted in significant differences in volume measurements for tiny nodules. A spiculated margin was shown to have a significant effect on nodule volume variability within a single size group. Use of different software packages resulted in significant volume measurement differences at the 2.5-mm CT slice thickness.
Keywords: chest • lung disease • MDCT • oncologic imaging • pulmonary nodules • volumetric evaluation and measurement
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Frequently, lung nodules are initially detected on screening CT examinations performed without thin-section technique, and follow-up CT studies may use varying techniques. For example, a nodule may be discovered on survey CT of the chest, abdomen, and pelvis using 5-mm-thick nonoverlapping sections, and the patient may return several months later for dedicated chest CT using 2.5-mm-thick overlapping sections. A number of effects of variations in scanning methods on volume measurements have been described in the literature [2-5]. The purpose of our study was to determine how slight differences in scanning collimation affect nodule volume measurements. Furthermore, we aimed to determine whether any differences in volume measurements were affected by nodule size, shape, or margination. In addition, we aimed to determine whether the use of two different commercially available software systems gave substantially different volume measurements.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Volumetric Analysis
Using GE Healthcare system—Volumetric measurements were
obtained on all three CT data set reconstructions for each nodule using Volume
Analysis software (GE Healthcare) on an Advantage workstation (GE Healthcare).
For each nodule, the volumes were labeled V1 (1.25-mm-thick sections), V2
(2.5-mm-thick sections), or V3 (5-mm-thick sections). Each individual
measurement was performed three times for the first 25 nodules with a
several-hour interval between measurements to assess intraobserver
variability; the remaining 50 nodules were measured only once for each data
set reconstruction.
Volumetric measurements were obtained on all 75 nodules on both the 1.25- and the 2.5-mm-thick sections. For 18 nodules, there was failure of segmentation on the 5-mm-thick sections and, as a result, no volumetric measurement was obtained on those nodules with that slice thickness.
Using R2 Technology-Vital Images system— Volumetric measurements for all 75 nodules were also attempted using LN500 R2 software (R2 Technology) on a Vitrea 2 workstation (Vital Images). Nodule segmentation using this software is not possible using 5-mm-thick sections; therefore, measurements were obtained using only 1.25- and 2.5-mm-thick CT sections. With this software, segmentation was successful in 71 of the 75 nodules on the 1.25-mm sections and 69 of the 75 nodules on the 2.5-mm sections. For six nodules, segmentation failed on the 2.5-mm sections; in four of these six nodules, segmentation also failed on the 1.25-mm sections. There were no nodules in which segmentation failed on the 1.25-mm sections without having failed on the 2.5-mm sections. For each nodule analyzed, the volumes were labeled V1* (1.25-mm-thick sections) or V2* (2.5-mm-thick sections).
Nodule Classification
The nodules were classified by a consensus of two radiologists according to
their size, shape, and margins based on visual analysis of the 1.25-mm-thick
axial images.
Size—The size of each nodule was characterized on the basis of its diameter into one of the following groups: tiny, diameter of 3-5 mm; small, diameter of 6-10 mm; medium, 11-15 mm; or large, diameter of 16-20 mm.
Shape—The shape of each nodule was classified as round (Fig. 1A) or elongated (Fig. 1B); elongated was defined as a nodule with a length that was > 1.5 x its width.
Margins—The margins of each nodule were characterized as smooth (including lobulated) (Fig. 2A) or spiculated (Fig. 2B).
Statistical Analysis
Analysis of measurements obtained on the GE Healthcare
system—The change in volume between volumetric measurements
obtained for the same nodule was calculated using the following formulas:
(V2 - V1)/V1, (V3 -
V2)/V2, and (V3 - V1)/V1. The
reference value used in all the calculations was the volumetric measurement
obtained using the thinner slice section of the two. A one-sample Student's
t test was used to determine whether these relative volume changes
were significantly different from zero.
Variability of volume measurements for each nodule was defined as follows: (|V1 - V2| + | V1 - V3|+| V2 - V3|)/(V1 + V2 + V3).
For those with missing V3 values, variability was defined as follows: |V1 - V2|/{(V1 + V2)/2}.
Because variability has a very skewed distribution, it was log-transformed for the comparison: log(variability + 0.001). Adding 0.001 was done to avoid log0, which is undefined.
Multiple linear regression was used to investigate the association of the logarithm of the volume variability (i.e., the dependent variable) with three independent variables: the nodule size (diameter), nodule shape, and nodule margination. The statistical software package used was S-plus (version 5.1, Insightful).
Two-sample Student's t tests were used to compare the variability between nodules with diameters of 3-10 mm (tiny and small nodules) and those with diameters of 11-20 mm (medium and large nodules).
Analysis of measurements obtained on the R2 Technology-Vital Images system—The change in volume for measurements obtained using different section thicknesses on the same nodule with the LN500 R2 software was calculated using the following formula: (V2* - V1*)/V1*.
Volume differences obtained from the two different software systems (Volume Analysis and LN500 R2) were calculated using the following formulas: (V1* - V1)/V1 and (V2* - V2)/V2. A one-sample Student's t test was used to determine whether these relative volume changes were significantly different from zero.
Variability in volume measurements for each individual nodule using the LN500 R2 software was calculated as follows:|V1* - V2*|/{(V1* + V2*)/2}.
Variability data were analyzed in the same fashion as the data obtained using the GE Healthcare system using multiple regression.
Two-sample Student's t tests were used to compare the variability between nodules with diameters of 3-10 mm (tiny and small nodules) and those with diameters of 11-20 mm (medium and large nodules).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Volume Analysis
GE Healthcare system—Repeating the volumetric measurement
calculation on the Volume Analysis software three times for each individual
nodule using each of the three slice thicknesses for the first 25 nodules
established 100% concordance for the individual measurements. Considering the
total lack of intraobserver variability, we did not repeat this process for
the remaining 50 nodules.
The mean volumes increased with increased section thickness as follows: V1, 348 mm3 (range, 19-2,200 mm3); V2, 351 mm3 (range, 16-2,200 mm3); and V3, 415 mm3 (range, 34-2,400 mm3). The mean volumes also showed mild increase with increasing section thickness when only the 57 nodules that could be segmented at all three CT section thicknesses were considered as follows: V1, 408 mm3 (range, 25-2,200 mm3); V2, 409 mm3 (range, 27-2,200 mm3); and V3, 415 mm3 (range, 34-2,400 mm3). However, individual nodule volumes sometimes increased and sometimes decreased with the use of thicker sections, as shown in Figures 3A, 3B, 3C.
|
|
Comparisons between V2 and V3 or V1 and V3 included only 57 nodules because automatic segmentation failed with 5-mm-thick sections for the remaining 18 nodules and, therefore, V3 was not obtainable in those cases. Those 18 nodules included 14 tiny and three small lesions and one large nodule. In 32 (56%) of 57 nodules, V3 was greater than V1 by up to 43%; in 21 (37%) nodules, V3 was less than V1 by up to 45%; and in four (7%), there was no change in volume. In 31 (54%) of 57 nodules, V3 was greater than V2 by up to 50%; in 22 (39%), V3 was less than V2 by up to 57%; and four (7%) nodules showed no change in volume.
Comparison of V1 with V2 revealed that 47 (63%) of 75 nodules showed < 10% change in volume (either increased or decreased), 15 (20%) showed 10-20% change in volume, and 13 (17%) showed > 20% in volume; a breakdown by nodule size is shown in Figure 4A. Comparison of both V2 with V3 and V1 with V3 showed that 27 (47%) of 57 nodules changed < 10% in volume, 14 (25%) changed 10-20% in volume, and 16 (28%) changed > 20% in volume; a breakdown by nodule size is shown in Figures 4B and 4C. Smaller nodules tended to show higher variation in calculated volumes compared with larger nodules. For tiny nodules, the relative volume change between V1 and V2 was calculated as [(V2 - V1)/V1], and the relative volume change between V1 and V3 was calculated as [(V3 - V1)/V1], both of which were significantly different from zero (p = 0.04 and 0.02, respectively). The relative volume change between V2 and V3 was not significant for tiny nodules. For small, medium, and large nodules, the relative volume changes between V1 and V2 and between V2 and V3 were not significant (p > 0.05).
|
|
|
|
|
The two-sample Student's t test comparing nodule volume variability between nodules 3-10 mm in diameter (tiny and small combined) and 11-20 mm in diameter (medium and large combined) showed that the logarithm of the variability in measurement of the tiny and small nodule group (mean = -2.211) was significantly larger than that of the medium and large nodule group (mean = -3.817), with a p value of < 0.0001.
R2 Technology-Vital Images system— There were 71 of 75 successfully segmented nodules using the LN500 R2 software at 1.25-mm CT sections and 69 nodules successfully segmented using the same software at both 1.25- and 2.5-mm CT sections. Only the 69 nodules that had successfully segmented at both slice thicknesses were considered in the analysis. This set of nodules included 29 tiny, 27 small, nine medium, and four large nodules. Eighteen of the 69 nodules were elongated, 51 were round, 55 were smooth, and 14 were spiculated.
The mean volume increased with increasing slice thickness: V1*, 390 mm3 (range, 18-2,231 mm3); and V2*, 396 mm3 (range, 14-2,458 mm3). Again, as with the Volume Analysis software, individual nodule volumes sometimes increased and sometimes decreased with the use of thicker CT sections. In 15 of 69 nodules, V2* was greater than V1* by up to 32%; in 52 nodules, V2* was less than V1 by up to 49%; and in two nodules, there was no change in volume.
Comparison of V1* and V2* revealed that 35 of 69 nodules showed < 10% change in volume (either increased or decreased), 19 nodules showed 10-20% change in volume, and 15 nodules showed a > 20% change in volume. Smaller nodules tended to show higher variation in calculated volumes than large nodules.
For tiny and small nodules, the relative volume changes between V1* and V2*, calculated as [(V2* - V1*)/V1*], were significantly different from zero (p < 0.0001 and 0.006, respectively). The relative volume changes between V1* and V2* were not significant for medium and large nodules (p >0.05).
The regression analysis results are shown in Table 4. Significantly increased nodule volume measurement variability was seen for tiny nodules compared with medium or with large nodules. No significant difference in volume measurement variability was noted when comparing tiny nodules with small nodules. In contrast to the data obtained using the Volume Analysis (GE Healthcare) software, nodule margination was not found to have a significant effect on nodule volume variability. Nodule shape was also found not to have a significant effect on nodule volume variability, as with the Volume Analysis data.
|
The two-sample Student's t test comparing nodule volume variability between nodules 3-10 mm in diameter (tiny and small combined) and 11-20 mm in diameter (medium and large combined) showed that the logarithm of the variability in measurement of the tiny and small nodule group (mean = -2.223) was significantly larger than that of the medium and large nodule group (mean = -3.470), with a p value of < 0.0001.
Comparison of the two systems—For one of the nodules, the measured volumes on the GE Healthcare system and R2 Technology-Vital Images system were so discrepant that this nodule was excluded from any comparison analysis (V1 = 143 mm3 and V2 = 166 mm3 vs V1* = 774 mm3 and V2* = 793 mm3).
Individual V2* versus V1* values (R2 Technology-Vital Images system) for these 68 nodules showed changes ranging from -49% to 32% (12% mean absolute change). Individual V2 versus V1 values (GE Healthcare system) showed changes of -22% to 52% (10% mean absolute change) in the same set of nodules. Individual V1* (R2 Technology-Vital Images system) versus V1 (GE Healthcare system) values showed differences ranging from -34% to 64%, with a mean absolute difference of 8%. The mean of relative nonabsolute differences between the two sets of measurements was only 0.2%, which was not significant (p = 0.9206). Individual V2* versus V2 values showed differences varying from -61% to 54%, with a mean absolute difference of 18%. The mean of relative nonabsolute differences between the two sets of measurements was -9.6%, which was significant (p = 0.0008).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our study concentrated on investigating the effect of CT slice thickness, in conjunction with nodule diameter, shape, and margination, on volume measurements. We also compared two different commercially available software systems for volumetric analysis. The study design, using the same scan data reconstructed with three different thicknesses, eliminated many of the potentially confounding factors described, such as differences in nodule location, degree of motion artifact, phase of cardiac cycle, or degree of respiration. Nodule segmentation and volumetric calculations were performed in an automated fashion, eliminating human bias in defining nodule margins. No juxtapleural nodules were included, thus eliminating the difficulty with automatic segmentation of such lesions. Whereas Revel and colleagues [9] found up to 6% variability in repeated volume measurements obtained from the same data set in 33% of their nodules, we found no intraobserver variability in any of the 25 nodules that we tested; this discrepancy may have been due to the large proportion of juxtapleural nodules in the data set used by Revel et al. compared with the total lack of juxtapleural nodules in our study.
In previously published studies, authors have reported greater variability and inaccuracy in volume determination of small nodules compared with large nodules for a given section thickness [2, 5, 8]. In one such study, tiny nodules (2-5 mm) varied 18.5% in volume, whereas large nodules (8-10 mm) varied 7.5% in volume [8]. In addition, recent investigations reported significant changes in volume measurements in a sizable proportion of nodules when different section thicknesses were used [4, 5]. In our study, we found that the degree of variability for small nodules was greater than the degree of variability for large nodules when comparing volumes obtained with different section thicknesses. Presumably, this finding is due to increased partial volume averaging for small nodules when thick sections are used; the effect of partial volume averaging is not as severe with large nodules. Partial volume averaging could also account for the failure of segmentation of 18 nodules on 5-mm CT sections, with 14 of these 18 nodules being 5 mm or less in greatest dimension. The results of our study suggest that thick sections (e.g., 5 mm) may be adequate for analysis of large nodules (> 15 mm in diameter), whereas thin sections (e.g., 1.25 mm) appear to be necessary for nodules < 10 mm.
In an investigation, Winer-Muram and colleagues [2] found that an increase in section thickness usually led to an increase in the estimation of nodule volume. However, that study was limited by manual, rather than automatic, nodule margin definition; the use of data sets obtained on two different dates, rather than using the same intrinsic scan data; and the use of thick sections (8 or 10 mm) in all but one patient. Zhao et al. [4] recently reported that increased section thickness led to increased mean calculated nodule volumes in lung metastases, although results for individual nodules were not reported. Our data also showed increased mean nodule volume with increased section thickness; however, analysis of individual volumes revealed that volume increased substantially in some nodules and decreased substantially in others. By reporting only mean volumetric data, it is possible that the findings reported by Zhao et al. may not reflect the full variation in volumetric results that was observed. Whereas our study showed lack of a statistically significant difference between volume measurements for several nodule size groups using different section thicknesses, many individual nodules showed large volume differences.
Neither the size breakdown nor the margination of nodules is specified in the study of Zhao et al. [4], so there may be differences in nodule populations between that study and ours. Moreover, those authors used a custommade volume system, whereas we used commercially available systems. In addition, section thicknesses used in our study (5, 2.5, and 1.25 mm) were thinner than those used by Zhao et al. [4] (7.5, 5, and 3.75 mm). Currently, CT for lung cancer screening and follow-up of small nodules that might represent bronchogenic carcinoma is generally performed using section thicknesses of no more than 2.5 mm, whereas screening and followup for lung metastases may be performed using thicker sections; thus, the methodology used to design these two studies differed because researchers had different patient populations in mind.
Winer-Muram and colleagues [2] found increased volume variability for irregular nodules, and our study showed similar results for spiculated versus smooth nodules, at least using one of the software packages. This is particularly important in lung cancer screening programs because spiculated nodules have traditionally been considered more suspicious for malignancy than smooth nodules.
Not only did our study use commercially available nodule segmentation software, but it also used two different software systems. The results indicate that although the effect of slice thickness is similar using both of these systems, there can be large differences in individual nodule volume measurements when shifting from one software system to another.
Our study had several limitations, including selection bias, with regard to nodules that were analyzed. Because juxtapleural, juxtavascular, and ground-glass nodules were excluded, our results cannot be applied to those types of commonly encountered nodules. In addition, our examinations lacked a small, focused field of view, possibly limiting the spatial resolution; a full patient-appropriate field of view was used because these were clinical examinations obtained to examine the lungs in their entirety. We had no proof of true nodule volumes; however, the purpose of our study was to evaluate the differences in calculated volumes using different section thickness, not to determine the accuracy of the volume technique. Because of the study design, many of the variables inherent in physically scanning a patient twice were eliminated; thus, our results reflect a best-case scenario with regard to measurement variability, and it would be expected that actual follow-up scanning would lead to larger apparent changes in nodule volumes.
In conclusion, although few statistically significant differences were
found between volume measurements obtained using different slice thicknesses,
our results indicate that a substantial proportion of individual nodules
showed a large change with different section thicknesses on both software
packages. This false change in nodule volume was most likely to occur for tiny
and small (
10 mm) nodules. When different software packages were used,
significantly different volume measurements were generated at a 2.5-mm CT
section thickness for all sizes of nodules. Nodule shape did not appear to
affect the results using either software package, whereas margination affected
volume measurements when using one of the software packages. The effect of
section thickness was also found to be different in the two software systems.
We therefore suggest than when following a nodule over time, it is best to use
the same section thickness and overlap and the same segmentation software for
all studies.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  O HONDA, M KAWAI, T GYOBU, Y KAWATA, T JOHKOH, J SEKIGUCHI, N TOMIYAMA, S YOSHIDA, H SUMIKAWA, A INOUE, et al. Reproducibility of temporal volume change in CT of lung cancer: comparison of computer software and manual assessment Br. J. Radiol., September 1, 2009; 82(981): 742 - 747. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Nietert, J. G. Ravenel, W. M. Leue, J. V. Miller, K. K. Taylor, E. S. Garrett-Mayer, and G. A. Silvestri Imprecision in Automated Volume Measurements of Pulmonary Nodules and Its Effect on the Level of Uncertainty in Volume Doubling Time Estimation Chest, June 1, 2009; 135(6): 1580 - 1587. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Gavrielides, L. M. Kinnard, K. J. Myers, and N. Petrick Noncalcified Lung Nodules: Volumetric Assessment with Thoracic CT Radiology, April 1, 2009; 251(1): 26 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wang, R. J. van Klaveren, H. J. van der Zaag-Loonen, G. H. de Bock, H. A. Gietema, D. M. Xu, A. L. M. Leusveld, H. J. de Koning, E. T. Scholten, J. Verschakelen, et al. Effect of Nodule Characteristics on Variability of Semiautomated Volume Measurements in Pulmonary Nodules Detected in a Lung Cancer Screening Program Radiology, August 1, 2008; 248(2): 625 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Bankier, C. R. O'Donnell, and P. M. Boiselle Quality Initiatives Respiratory Instructions for CT Examinations of the Lungs: A Hands-on Guide1 RadioGraphics, July 1, 2008; 28(4): 919 - 931. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. Ravenel, W. M. Leue, P. J. Nietert, J. V. Miller, K. K. Taylor, and G. A. Silvestri Pulmonary Nodule Volume: Effects of Reconstruction Parameters on Automated Measurements--A Phantom Study Radiology, May 1, 2008; 247(2): 400 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Suzuki, H. Jacobsson, T. Hatschek, M. R. Torkzad, K. Boden, Y. Eriksson-Alm, E. Berg, H. Fujii, A. Kubo, and L. Blomqvist Radiologic Measurements of Tumor Response to Treatment: Practical Approaches and Limitations RadioGraphics, March 1, 2008; 28(2): 329 - 344. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
