Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2353040737
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Goo, J. M.
Right arrow Articles by Bae, K. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Goo, J. M.
Right arrow Articles by Bae, K. T.
(Radiology 2005;235:850-856.)
© RSNA, 2005

Experimental Studies

Volumetric Measurement of Synthetic Lung Nodules with Multi–Detector Row CT: Effect of Various Image Reconstruction Parameters and Segmentation Thresholds on Measurement Accuracy1

Jin Mo Goo, MD, Trongtum Tongdee, MD, Ranista Tongdee, MD, Kwangjae Yeo, MS, Charles F. Hildebolt, DDS, PhD and Kyongtae T. Bae, MD, PhD

1 From the Mallinckrodt Institute of Radiology, Washington University School of Medicine, 510 S Kingshighway Blvd, St Louis, MO 63110 (J.M.G., T.T., R.T., K.Y., C.F.H., K.T.B.); and Department of Radiology, Seoul National University College of Medicine, and the Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Korea (J.M.G.). Received April 23, 2004; revision requested July 2; revision received July 19; accepted August 18. Address correspondence to K.T.B. (e-mail: baet@mir.wustl.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To evaluate the effect of various multi–detector row computed tomographic (CT) reconstruction parameters and nodule segmentation thresholds on the accuracy of volumetric measurement of synthetic lung nodules.

MATERIALS AND METHODS: Synthetic lung nodules of four different diameters (3.2, 4.8, 6.4, and 12.7 mm) were scanned with multi–detector row CT. Images were reconstructed at various section thicknesses (0.75, 1.0, 2.0, 3.0, and 5.0 mm), fields of view (30, 20, and 10 cm), and reconstruction intervals (0.5, 1.0, and 2.0 mm). The nodules were segmented from the simulated background lung region by using four segmentation thresholds (–300, –400, –500, and –600 HU), and their volumes were estimated and compared with a reference standard (measurements according to fluid displacement) by computing the absolute percentage error (APE). APE was regressed against nodule size, and multivariate analysis of variance (MANOVA) was performed with APE as the dependent variable and with four within-subject factors (field of view, reconstruction interval, threshold, and section thickness).

RESULTS: The MANOVA demonstrated statistically significant effects for threshold (P = .02), section thickness (P < .01), and interaction of threshold and section thickness (P = .04). The regression of mean APE values on nodule size indicates that APE progressively increases with decreasing synthetic nodule size (R2 = 0.99, P < .01).

CONCLUSION: For accurate measurement of lung nodule volume, it is critical to select a section thickness and/or segmentation threshold appropriate for the size of a nodule.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
The sensitivity for detecting small lung nodules has increased substantively with thin-section multi–detector row computed tomography (CT) (1). Small lung nodules, many of which are found incidentally, pose a serious management challenge, particularly for the practical implementation of CT-based lung cancer screening. Despite the availability of sophisticated schemes for automated detection and characterization of lung nodules (2–5), it is difficult to determine on the basis of morphology whether small nodules are malignant or benign (6–8).

One of the widely accepted characteristics of nodule malignancy is growth. The evaluation of nodule growth, however, is not without challenge. A commonly used method based on the linear measurement of nodule diameters is often inconsistent and unreliablefor detecting volumetric changes (9,10). Because nodules that are considered malignant would be further evaluated or treated more aggressively, it is important to obtain accurate and reliable measurements of nodule volume changes and thus increase diagnostic confidence. Consequently, with increasing applications of thin-section CT and lung cancer screening, the assessment of nodule growth has been an active area of research (11–16). Computer-aided volumetric assessment of small pulmonary nodules for estimating growth is gaining wider acceptance (11,12,15).

The acquisition of appropriate three-dimensional imaging data is a prerequisite for accurate and reliable nodule volume measurement. There are many scanning and reconstruction parameters that affect the quantification of nodule volume (17,18). In general, improvement of spatial resolution by means of reduction of section thickness, field of view (FOV), and reconstruction interval should decrease errors in nodule volume measurement. The relative importance of these reconstruction parameters on the measurement of nodule volume, however, is not equal and likely depends on the size of nodules. Furthermore, with three-dimensional measurement methods, the nodules are segmented from the background lung parenchyma on CT images prior to volume measurement. The segmentation is usually based on gray-level thresholding, and the choice of threshold values influences the measured volumes of the segmented nodules. Although authors of several studies have investigated the effects of nodule size (13,14,19), thresholds (13,19), and section thickness (14) on the volumetric measurement of lung nodules, the combined effects of these parameters have not been reported. Thus, the purpose of our study was to evaluate the effects of various multi–detector row CT reconstruction parameters and nodule segmentation thresholds on the accuracy of volumetric measurement of synthetic lung nodules.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Synthetic Lung Nodules and the Reference Standard
Acrylic spheres of four different diameters (3.2, 4.8, 6.4, and 12.7 mm) were used to simulate lung nodules. The size of our synthetic small nodules was selected such that the lower size limit was 3 mm, because this is commonly considered tobe the lowest detection limit on clinical lung CT images. The upper size limit of a small nodule was chosen to be 13 mm, which is slightly smaller than half of the upper limit of a nodule as defined by the Fleischner Society (20). The mean attenuation value of these spheres was approximately 130 HU, which is recognized as higher than that of clinically noncalcified soft-tissue nodules (30–60 HU). To obtain the reference-standard volume of the nodules, we weighed the nodule on an analytical chemistry balance, with a precision of 0.1 mg. The reference-standard volume of a nodule was calculated by multiplying the measured weight by the specific gravity. The specific gravity of a nodule was computed from the weight of the largest nodule divided by the volume that was determined by using water displacement. This approach was based on our observation that the volume measurement of the largest nodule was less variable than that of the smaller nodules obtained with water displacement. The weights and volume measurements were repeated three to four times and were averaged to calculate the values used in our analyses. The apparent diameter of each nodule was calculated from the estimated nodule volume with the assumption that the nodules were spherical in shape.

A chest CT phantom (Computerized Image Reference Systems, Norfolk, Va) was used to simulate a 5-cm-thick transverse section of the thorax (Fig 1). The lung region of the chest CT phantom was filled with a sculpted slab of polystyrene plastic matrix material (mean attenuation value, –990 HU). The synthetic nodules were embedded in this matrix material.



View larger version (103K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Transverse CT scan of chest phantom with eight synthetic nodules embedded in a polystyrene plastic matrix occupying the lung region. Only nodules in the right lung were measured for this study because images with FOV of 10 and 20 cm could only cover the right lung.

 
Imaging Protocols
The phantom was placed on a CT table with its transverse imaging plane (Fig 1) perpendicular to the direction of table motion. The scans were obtained by using a 16–detector row CT scanner (Somatom Sensation 16; Siemens, Forchheim, Germany) with a 16 x 0.75-mm collimation, 120 kVp, 255 mA, and 500-msec gantry rotation time in a spiral mode. Several image sets were reconstructed from the raw data by choosing five different section thicknesses (0.75, 1.0, 2.0, 3.0, and 5.0 mm) and three FOVs (30, 20, and 10 cm). The FOVs of 30, 20, and 10 cm correspond to in-plane resolution of 0.59, 0.39, and 0.20 mm, respectively. The images were reconstructed with three intervals (0.5, 1.0, and 2.0 mm) by using a lung image reconstruction kernel (B60f sharp).

Nodule Volume Measurement on CT Images
All CT image sets were transferred to a personal computer with a 2.8-GHz Pentium IV processor (Intel, Santa Clara, Calif) and 1 GB of RAM. The images were displayed and processed by using a commercially available software program (Rapidia; Infinitt, Seoul, Korea). The program allowed the user to segment and measure the volume of the nodules at a user-defined attenuation threshold. After a nodule was identified and clicked on the screen by using a computer mouse, it was segmented in a fully automated fashion on the gray-level thresholds. All of the nodule volume measurements were performed by one radiologist (J.M.G., 13 years of experience with chest CT). We used four thresholds: –300, –400, –500, and –600 HU. These values were selected arbitrarily around the mean attenuation (–430 HU) of the synthetic nodule (130 HU) and lung (–990 HU) attenuations.

A total of 720 nodule volume measurements (five section thicknesses x three FOVs x three reconstruction intervals x four thresholds x four nodules) were evaluated.

Data and Statistical Analyses
For each CT image-based nodule volume measurement, the absolute percentage error (APE) was calculated as 100 x |Vm – Vrs|/Vrs, where Vm and Vrs are the CT measured and reference-standard nodule volumes, respectively.

Mean APE values were calculated for the four sizes of synthetic nodules. Regression analysis was used to assess the association between the mean APE value and nodule size. A multivariate analysis of variance was performed with four within-subject factors (FOV, reconstruction interval, threshold, and section thickness). Interaction terms were included in the analysis. The Mauchley test was used to evaluate the sphericity assumption. To help assess interactions, the percentage error was plotted against nodule diameter, FOV, reconstruction interval, threshold, and section thickness. Statistical analyses were performed with Statistica software (version 6.0; StatSoft, Tulsa, Okla) and JMP statistical software (version 5.1; SAS, Cary, NC). P < .05 indicated a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
The reference-standard volumes of the four nodules were estimated to be 16.5, 56.6, 135.0, and 1072.5 mm3, which corresponded to apparent spherical diameters of 3.2, 4.8, 6.4, and 12.7 mm, respectively.

We were unable to segment and measure the volume in 13 of 720 nodule combinations, because these nodules were faintly delineated because of partial volume effects. All of the missing values occurred with nodule diameters of 3.2 mm. Five values occurred at FOVs of 30 cm, four occurred at FOVs of 20 cm, and four occurred at FOVs of 10 cm. Seven missing values occurred at reconstruction intervals of 2.0 mm, three occurred at reconstruction intervals of 1.0 mm, and three occurred at reconstruction intervals of 0.5 mm. Ten of the missing values occurred at thresholds of –300 HU and three occurred at thresholds of –400 HU. Twelve missing values occurred at a section thickness of 5.0 mm and one occurred at a section thickness of 3.0 mm. Because we could not segment these nodules, our best estimate of their volumes was zero. The APEs for these measurements were, therefore, 100.

The multivariate analysis of variance demonstrated statistically significant effects for threshold (P = .02), section thickness (P < .01), and the interaction of threshold and section thickness (P = .04). Findings of Mauchley tests of sphericity were nonsignificant for these effects (P ≥ .06); therefore, the reported P values are for the univariate F test. FOV, reconstruction interval, and other interactions (second, third, and fourth degree) were not statistically significant (P > .05).

Plots of mean APEs versus nodule sizes for five different section thicknesses are presented in Figure 2. Similar plots for four different thresholds are presented in Figure 3. The plots indicate consistently increasing APE with decreasing nodule size except for the threshold value of –600 HU for the 3.2-mm nodule size. To understand this exception, APEs for a 3.2-mm nodule and at threshold of –600 HU were tabulated (Tables 1, 2).



View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Graph shows the mean APE versus nodule size for five different section thicknesses. At each section thickness, APE increased consistently with a decrease in nodule size.

 


View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph shows the mean APE versus nodule size for four different thresholds. At each threshold, except for the threshold value of –600 HU and 3.2-mm nodule, APE increased consistently with a decrease in nodule size.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Mean APE Measured in a 3.2-mm Nodule

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Mean APE at Threshold of –600 HU

 
Regression analysis was used to assess the overall association between nodule size and APE. To do this, grand means of pooled data were calculated for APE values, and log transformations of the data were performed to fit the data in a linear regression model (21). The resulting regression of mean APE values on synthetic nodule sizes was highly significant (R2 = 0.99, P < .01).

On scatter plots of percentage errors against nodule diameter, FOV, reconstruction interval, threshold, and section thickness, nodule volume tends to be overestimated with a threshold of –600 HU and underestimated with thresholds of –300 and –400 HU. As the size of a nodule decreased, the range of error increased in both directions (Fig 4).



View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Scatter plots show relationships between the percentage error and nodule diameter, FOV, reconstruction interval, threshold, and section thickness. The nodule volume was overestimated at a threshold of –600 HU but was underestimated at thresholds of –300 and –400 HU. As the size of nodules decreases, errors increase in both positive and negative directions.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
Because multi–detector row CT facilitates the acquisition and analysis of three-dimensional CT images, the use of volumetric analysis of lung nodules is likely to increase. The primary clinical application of nodule volumetric measurements is to monitor nodule growth (11,12). The three-dimensional measurements have obvious advantages over conventional bilinear measurements for accurately representing volumes. Authors (12) of a study about three-dimensional estimation of nodule volume reported that the volume of nodules as small as 3 mm in diameter could be measured with a high accuracy.

An image-based nodule volume measurement method, however, has many intrinsic and practical limitations and is subject to measurement errors. For correct interpretation of nodule growth, it is critical to understand how various parameters affect nodule volume measurements and become sources of measurement error. Ko et al (13) evaluated the effect of imaging variables on the volumetric measurement of lung nodules by using a realistic phantom and suggested that a partial volume method, a high-frequency reconstruction algorithm, and CT performed with a diagnostic dose (120 mAs) would improve the precision of the volume measurements of lung nodules. Winer-Muram et al (14) measured and analyzed the volumes of simulated nodules in phantoms and in patients with stage I lung cancer and reported that the lung nodule volumes in phantoms were overestimated; this overestimation was directly proportional to the image section thickness and inversely proportional to the nodule size.

In the current study, we analyzed the effect of four key parameters (threshold, section thickness, FOV, and reconstruction interval) on nodule volume measurements by using synthetic lung nodules. Our results demonstrated the relative significance of these parameters with regard to how they affect measurement error. We believe this information contributes to the optimization of imaging protocols for accurate and repeatable nodule volume measurements with serial CT studies.

It is well known that varying the window center alters the measured volume even for relatively large lesions or nodules (9,19,22). Our study findings also revealed that the segmentation threshold is a significant determinant of measurement errors. The most common approach for measuring the volume of lung nodules is based on gray-level thresholding, which allows a user to segment lung nodules from the background parenchyma and measure the volume of the segmented voxels (12,13,15,16). The selection of an appropriate threshold for lung nodule segmentation is crucial. Optimal threshold levels may be determined from valleys on the histogram of lung nodule and parenchyma attenuations or from levels that provide a statistical maximal separation between the regions to be distinguished (16). Previous study findings have demonstrated that a structure is most accurately represented when the display window center is set midway between the CT attenuation of the structure of interest and the background (22). In our study, thresholds of –500 and –400 HU, which are closer to the mean of the attenuations of the nodule (130 HU) and the background (–990 HU), resulted in smaller errors than thresholds of –600 and –300 HU (Fig 3). In practice, the application of a single fixed threshold is the simplest and most consistent approach for segmenting nodules. The midway attenuation between the nodules and parenchyma may not be fixed, however, because the attenuation values of nodules vary depending on their type. Furthermore, segmentation for nonsolid and part-solid nodules frequently requires a much lower threshold than that required for solid nodules. Even in the same nodule, small peripheral lung neoplasms may have a replacement growth pattern. As found in a study by Kakinuma et al (23), the ground-glass opacities of lung cancer nodules not only increase in size or attenuation but may also decrease rapidly or slowly with the appearance of solid components. This variable growth pattern could not be recognized with the use of a single fixed threshold.

The results of our study also demonstrated that section thickness is an important parameter in nodule volume measurement. With multi–detector row CT, thin-section volumetric imaging in the whole lung becomes routine, particularly for lung cancer screening studies (24,25). A drawback of this practice is that a thin-section CT scan of the whole thorax generates a large dataset (typically 250–350 images of 1.0-mm section thickness), which requires considerable time to interpret and analyze (26).

With the CT image matrix size fixed at 512 x 512, the voxel dimension is proportional to the FOV. Some studies used a FOV of 9.6 cm (11,12) in measuring nodule volumes, while another study used a 30-cm FOV (13). The effect of FOV on volume measurement has not, however, been studied. In the current study, FOV did not have a significant effect on the APE. This result is not surprising, in part, because the size of in-plane resolution differences (0.20–0.59 mm) due to the choices in FOV in our study was much lower than that of longitudinal resolution differences (0.75–5.0 mm) due to the choices in section thickness.

In three-dimensional imaging, the reconstruction interval is one of the parameters that affects the longitudinal resolution. With smaller reconstruction intervals, the number of overlapping transverse images increases and the longitudinal resolution improves. Brink et al (27) suggested that single–detector row helical CT images should be reconstructed with at least 60% overlap relative to the effective section thickness to obtain maximal longitudinal resolution. It is possible that a further increase in the overlap percentage may not provide a measurable gain in the longitudinal resolution. Although reconstruction interval was not expressed as a percentage of section thickness in our study, reconstruction interval itself was not a significant parameter in determining volume measurement error.

For a 3.2-mm nodule imaged on 5.0- or 3.0-mm-thick sections, the attenuation of all the pixels within the nodule was below –300 HU in 10 cases and –400 HU in three cases because of the partial volume effect. Therefore, the volumes of these nodules were measurable at low thresholds (–500 or –600 HU) but not at high thresholds (–300 or –400 HU). We could not measure the volume of nodules in 2% of the cases. For these cases, we considered our measurement error to be 100%. We believe that this was a reasonable procedure for dealing with these cases.

In plots of error measurement against nodule size for thresholds and section thicknesses, the error progressively increased with decreasing nodule size except for the –600 HU threshold for the 3.2-mm nodule size. This phenomenon may have been caused by an interaction between parameters. Our statistical analysis demonstrated significant interaction between section thickness and threshold. Tables 1 and 2 were used to interpret this exception. In general, the error increases as the section thickness increases, and the nodule volume tends to be overestimated as the threshold decreases (Figs 2, 4). For a 3.2-mm nodule at a threshold of –600 HU, however, the error decreases as the section thickness increases (Tables 1, 2). The attenuation of the surface voxels of a nodule decreases as the section thickness increases because of partial volume effect, but at the same time the volume of the voxel also increases as the section thickness increases. The volume measurement error, therefore, depends on both the threshold and the section thickness; with some combinations, both parameters act in the same direction, and with other combinations, the parameters act in opposition direction. Compared with large nodules, small nodules are sensitive to these effects because the proportion of surface voxels is larger in small nodules. For a 3.2-mm nodule, thick section thickness acted in the direction of underestimation, which counteracted the effect of overestimation by a threshold of –600 HU.

Our analysis also demonstrated a statistically significant decreasing relationship between measurement error and nodule size: Smaller nodules tended to have higher measurement errors. As nodules become smaller, the partial volume effect and measurement error increase (12,14,15). This effect is reduced with the use of smaller voxel or isotropic voxel size.

There were several limitations of our study. We used motion-free spherical homogeneous nodule phantoms, which do not accurately represent clinical reality. All of the synthetic nodules had the same attenuation value, and nodules with attenuation equivalent to ground-glass opacity of lesions were not included. The software package we used was not specifically designed for measuring lung nodule volume, and the measurement of nodule volumes was based on a thresholding method. We, however, believe that this simplified phantom study design has allowed us to achieve the aim of our study, that is, the evaluation of the effect of reconstruction and technical parameters on nodule volume measurement error.

In summary, the results of our study demonstrated that nodule size, section thickness, and threshold were significant parameters in determining volume measurement errors. The FOV and reconstruction interval did not significantly affect volume measurement error. For an accurate volumetric measurement of a lung nodule, it is critical to select a section thickness and/or segmentation threshold appropriate for the size of the nodule.

Practical application: Accurate measurement of nodule volume is crucial in the current scheme of imaging-based assessment of malignancy in small nodules. Because the acquisition and reconstruction parameters for CT imaging can significantly affect the accuracy of nodule volume measurement, it is important to select appropriate parameters to achieve a desired accuracy. The results of the present study suggest that we should be more attentive to the selection of section thickness and segmentation threshold than to selection of FOV and reconstruction interval for an accurate measurement of nodule volume.


   FOOTNOTES
 
Abbreviations: APE = absolute percentage error, FOV = field of view

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, J.M.G., K.T.B.; study concepts, J.M.G., K.Y., K.T.B.; study design, J.M.G., T.T., R.T., C.F.H., K.T.B.; literature research, J.M.G., K.T.B.; experimental studies, J.M.G., T.T., R.T., K.Y.; data acquisition, J.M.G., T.T., R.T.; data analysis/interpretation, all authors; statistical analysis, J.M.G., C.F.H., K.T.B.; manuscript preparation, J.M.G., T.T., K.Y.; manuscript definition of intellectual content, J.M.G., R.T., C.F.H., K.T.B.; manuscript editing, J.M.G., K.T.B.; manuscript revision/review and final version approval, all authors


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Fischbach F, Knollmann F, Griesshaber V, Freund T, Akkol E, Felix R. Detection of pulmonary nodules by multislice computed tomography: improved detection rate with reduced slice thickness. Eur Radiol 2003; 13:2378-2383.[CrossRef][Medline]
  2. Armato SG, III, Li F, Giger ML, MacMahon H, Sone S, Doi K. Lung cancer: performance of automated lung nodule detection applied to cancers missed in a CT screening program. Radiology 2002; 225:685-692.[Abstract/Free Full Text]
  3. Brown MS, Goldin JG, Suh RD, McNitt-Gray MF, Sayre JW, Aberle DR. Lung micronodules: automated method for detection at thin-section CT—initial experience. Radiology 2003; 226:256-262.[Abstract/Free Full Text]
  4. Goo JM, Lee JW, Lee HJ, Kim S, Kim JH, Im JG. Automated lung nodule detection at low-dose CT: preliminary experience. Korean J Radiol 2003; 4:211-216.[Medline]
  5. Armato SG, III, Altman MB, Wilkie J, et al. Automated lung nodule classification following automated nodule detection on CT: a serial approach. Med Phys 2003; 30:1188-1197.[CrossRef][Medline]
  6. Munden RF, Pugatch RD, Liptay MJ, Sugarbaker DJ, Le LU. Small pulmonary lesions detected at CT: clinical importance. Radiology 1997; 202:105-110.[Abstract/Free Full Text]
  7. Henschke CI, McCauley DI, Yankelevitz DF, et al. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999; 354:99-105.[CrossRef][Medline]
  8. Goo JM, Chung MJ, Lee HJ, Im JG. Posterior subpleural nodules in patients with underlying malignancies: value of the prone computed tomography. J Comput Assist Tomogr 2003; 27:274-278.[CrossRef][Medline]
  9. Harris KM, Adams H, Lloyd DC, Harvey DJ. The effect on apparent size of simulated pulmonary nodules of using three standard CT window settings. Clin Radiol 1993; 47:241-244.[CrossRef][Medline]
  10. Erasmus JJ, Gladish GW, Broemeling L, et al. Interobserver and intraobserver variability in measurement of non-small-cell carcinoma lung lesions: implications for assessment of tumor response. J Clin Oncol 2003; 21:2574-2582.[Abstract/Free Full Text]
  11. Yankelevitz DF, Gupta R, Zhao B, Henschke CI. Small pulmonary nodules: evaluation with repeat CT—preliminary experience. Radiology 1999; 212:561-566.[Abstract/Free Full Text]
  12. Yankelevitz DF, Reeves AP, Kostis WJ, Zhao B, Henschke CI. Small pulmonary nodules: volumetric determined growth rates based on CT evaluation. Radiology 2000; 217:251-256.[Abstract/Free Full Text]
  13. Ko JP, Rusinek H, Jacobs EL, et al. Small pulmonary nodules: volume measurement at chest CT—phantom study. Radiology 2003; 228:864-870.[Abstract/Free Full Text]
  14. Winer-Muram HT, Jennings SG, Meyer CA, et al. Effect of varying CT section width on volumetric measurement of lung tumors and application of compensatory equations. Radiology 2003; 229:184-194.[Abstract/Free Full Text]
  15. Kostis WJ, Reeves AP, Yankelevitz DF, Henschke CI. Three-dimensional segmentation and growth-rate estimation of small pulmonary nodules in helical CT images. IEEE Trans Med Imaging 2003; 22:1259-1274.[CrossRef][Medline]
  16. Zhao B, Yankelevitz D, Reeves A, Henschke CI. Two-dimensional multi-criterion segmentation of pulmonary nodules on helical CT images. Med Phys 1999; 26:889-895.[CrossRef][Medline]
  17. Zerhouni EA, Spivey JF, Morgan RH, Leo FP, Stitik FP, Siegelman SS. Factors influencing quantitative CT measurements of solitary pulmonary nodules. J Comput Assist Tomogr 1982; 6:1075-1087.[Medline]
  18. Im JG, Gamsu G, Gordon D, et al. CT densitometry of pulmonary nodules in a frozen human thorax. AJR Am J Roentgenol 1988; 150:61-66.[Abstract/Free Full Text]
  19. Van Hoe L, Haven F, Bellon E, et al. Factors influencing the accuracy of volume measurements in spiral CT: a phantom study. J Comput Assist Tomogr 1997; 21:332-338.[CrossRef][Medline]
  20. Austin JH, Muller NL, Friedman PJ, et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology 1996; 200:327-331.[Free Full Text]
  21. Kleinbaum DG, Kupper LL, Muller KE. Applied regression analysis and other multivariate methods 2nd ed. Boston, Mass: PWS-Kent, 1987; 220.
  22. Baxter BS, Sorenson JA. Factors affecting the measurement of size and CT number in computed tomography. Invest Radiol 1981; 16:337-341.[CrossRef][Medline]
  23. Kakinuma R, Ohmatsu H, Kaneko M, et al. Progression of focal pure ground-glass opacity detected by low-dose helical computed tomography screening for lung cancer. J Comput Assist Tomogr 2004; 28:17-23.[CrossRef][Medline]
  24. Mahesh M. Search for isotropic resolution in CT from conventional through multiple-row detector. RadioGraphics 2002; 22:949-962.[Abstract/Free Full Text]
  25. Aberle DR, Gamsu G, Henschke CI, Naidich DP, Swensen SJ. A consensus statement of the Society of Thoracic Radiology: screening for lung cancer with helical computed tomography. J Thorac Imaging 2001; 16:65-68.[CrossRef][Medline]
  26. Rubin GD. Data explosion: the challenge of multidetector-row CT. Eur J Radiol 2000; 36:74-80.[CrossRef][Medline]
  27. Brink JA, Wang G, McFarland EG. Optimal section spacing in single-detector helical CT. Radiology 2000; 214:575-578.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 RadiologyHome page
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]

Home page
 RadioGraphicsHome page
A. A. Bankier, C. R. O'Donnell, and P. M. Boiselle
Quality Initiatives * Respiratory Instructions for CT Examinations of the Lungs: A Hands-on Guide
RadioGraphics, July 1, 2008; 28(4): 919 - 931.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
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]

Home page
 RadiologyHome page
O. Honda, T. Johkoh, H. Sumikawa, A. Inoue, N. Tomiyama, N. Mihara, Y. Fujita, M. Tsubamoto, M. Yanagawa, T. Daimon, et al.
Pulmonary Nodules: 3D Volumetric Measurement with Multidetector CT Effect of Intravenous Contrast Medium
Radiology, December 1, 2007; 245(3): 881 - 887.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
H Bolte, C Riede, S Muller-Hulsbeck, S Freitag-Wolf, G Kohl, T Drews, M Heller, and J Bieder
Precision of computer-aided volumetry of artificial small solid pulmonary nodules in ex vivo porcine lungs
Br. J. Radiol., June 1, 2007; 80(954): 414 - 421.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
M. Petrou, L. E. Quint, B. Nan, and L. H. Baker
Pulmonary Nodule Volumetric Measurement Variability as a Function of CT Slice Thickness and Nodule Morphology
Am. J. Roentgenol., February 1, 2007; 188(2): 306 - 312.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
S. C. Yeshwant, R. M. Summers, J. Yao, D. S. Brickman, J. R. Choi, and P. J. Pickhardt
Polyps: Linear and Volumetric Measurement at CT Colonography
Radiology, December 1, 2006; 241(3): 802 - 811.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J.-S. Kim, J.-H. Kim, G. Cho, and K. T. Bae
Automated Detection of Pulmonary Nodules on CT Images: Effect of Section Thickness and Reconstruction Interval--Initial Results
Radiology, July 1, 2005; 236(1): 295 - 299.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Goo, J. M.
Right arrow Articles by Bae, K. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Goo, J. M.
Right arrow Articles by Bae, K. T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE