(Radiology. 2000;215:175-183.)
© RSNA, 2000
Lung Cancer Screening: Minimum Tube Current Required for Helical CT1
Shigeki Itoh, MD,
Mituru Ikeda, MD,
Shoji Arahata, MD,
Takeshi Kodaira, MD,
Takayuki Isomura, MD,
Takashi Kato, MD,
Kouji Yamakawa, MD,
Kunihiro Maruyama, MD and
Takeo Ishigaki, MD
1 From the Departments of Radiology (S.I., S.A., T. Kodaira, T. Isomura, K.M., T. Ishigaki) and Medical Information and Medical Records (M.I.), Nagoya University School of Medicine, 65 Turumai-cho Showa-ku, 466-0065 Nagoya, Japan; the Department of Biofunctional Research, National Institute for Longevity Science, Oubu, Japan (T. Kato); and the Department of Radiology, Tosei General Hospital, Seto, Japan (K.Y.). From the 1998 RSNA scientific assembly. Received November 10, 1998; revision requested January 21, 1999; final revision received September 1; accepted September 9. Address reprint requests to S.I.
 |
Abstract
|
|---|
PURPOSE: To investigate the minimum tube current required for helical computed tomography (CT) for lung cancer screening.
MATERIALS AND METHODS: Thirty helical scans of the lung were obtained at effective tube currents of 50, 30, 20, 18, 12, 10, and 6 mAs in seven healthy volunteers. Computer-generated nodules 6 mm in diameter that showed ground-glass opacity were superimposed on the images. The image quality and detectability of nodules were evaluated subjectively by six observers. The SDs of measured CT numbers were calculated. The results were analyzed according to location in the lung.
RESULTS: Compared with the subjective quality of images obtained at 50 mAs, the subjective quality of images obtained at 20 mAs was not significantly different. The detectability of nodules was not significantly degraded by reducing the tube current to 20 mAs in the upper zone of the lung, to 12 mAs in the middle zone, or to 18 mAs in the lower zone. The SDs at the apex and base of the lung were larger than those at other levels, and the difference became greater as the dose was reduced.
CONCLUSION: The minimum tube current required for screening helical CT differs for different locations in the lung. An ideal CT protocol for the lung should permit the tube current to be changed during helical scanning.
Index terms: Cancer screening, 60.12115, 60.32 • Computed tomography (CT), helical, 60.12115 • Computed tomography (CT), image quality, 60.12115 • Lung neoplasms, CT, 60.12115, 60.32 • Lung neoplasms, diagnosis, 60.12115, 60.32
|
Introduction
|
|---|
Lung cancer is now the leading cancer cause of death in Japan (1). The prognosis is poor because early curable lung cancers usually produce no symptoms and are often missed in mass screening programs for lung cancer in which chest radiography is employed (2). The cure rate for non–small cell lung cancers less than 20 mm in diameter has been reported to be greater than 80% (1). Therefore, earlier and more certain detection with more effective screening methods can be expected to improve cure rates.
Recently, attempts have been made in Japan to apply helical computed tomography (CT) to lung cancer screening (3,4). The preliminary results in populations at high risk for developing lung cancer showed that helical CT is a potentially more useful screening method for the detection of early peripheral lung cancers than is chest radiography (4). However, since the radiation dose associated with this method, in which helical CT is performed at a tube current of 50 mAs and at a pitch of 2.0, is about 10 times higher than that associated with conventional chest radiography, a further reduction in radiation dose is required before this method can be applied to the general population.
There is still controversy concerning the minimum tube current required for CT of the chest. Naidich et al (5) reported that high-quality diagnostic images of the lung can be obtained with conventional CT at 20 mAs, and Zwirewich et al (6) also concluded that thin-section CT images acquired at 20 mA yield anatomic information equivalent to that obtained at 200 mA without a substantial loss in spatial resolution or degradation of image quality. On the other hand, Mayo et al (7) reported that 140 mAs is the minimum tube current required to provide good image quality in conventional CT.
To our knowledge, the limits of low-dose helical CT for lung cancer screening have not been evaluated in any study. Therefore, we conducted this basic study to investigate the minimum tube current required for helical CT in mass screening for lung cancer.
|
MATERIALS AND METHODS
|
|---|
The study protocol was approved by the ethics committee of our institution, and informed consent was obtained from all subjects after the purpose and protocol of this study had been fully explained. The study group comprised seven healthy volunteers (six men, one woman; age range, 38–59 years; mean age, 50 years).
All images were acquired by using a Vigor helical CT scanner (Toshiba, Tokyo, Japan). Helical CT was performed with the fixed scanning parameters of 120 kVp, 1 second per rotation, 10-mm collimation, and a pitch of 2.0 (table feed rate, 20 mm/sec), and the entire lung was scanned during a single breath hold at maximum inspiration. Images were reconstructed at 10-mm intervals with 180° linear interpolation by using lung window settings (window width, 1,600 HU; window level, -600 HU). These are the same scanning parameters employed in helical CT for lung cancer screening at most institutions in Japan (4). Mediastinal settings were not included in this study.
Thirty helical scans of the lung were obtained at effective tube currents of 50, 30, 20, 18, 12, 10, and/or 6 mAs in the seven healthy volunteers (Table 1). We used a tube current of 50 mAs for all volunteers as a standard of reference and assigned lower tube currents to each patient at random, but the number of scans for each tube current was intended to be the same. Effective tube currents of 18, 12, and 6 mAs were obtained by using an aluminum filter designed for CT fluoroscopy (8). Because in our preliminary study (9) this filter provided a reduction of 40% in the absorbed dose compared with the standard scanning dose, scanning performed at a tube current of 30 mAs with this filter corresponds to a tube current of 18 mAs without this filter. In this study, a tube current of 50 mAs was used for comparison because findings of previous studies have demonstrated that the image quality at this tube current is sufficient for helical CT lung cancer screening (3,4,10,11) and because this allowed us to reduce the radiation dose in volunteers markedly compared with the dose at 200 mAs.
View this table:
[in this window]
[in a new window]
|
TABLE 1. Effective Tube Current Used in Examinations and Cross-sectional Diameter of the Base of the Lung in Healthy Volunteers
|
|
We synthesized computer-generated nodules of about 6 mm in diameter by increasing the CT number and superimposed these simulated nodules onto the images obtained in the healthy volunteers; the increases in the CT number in these nodules over the CT number of the background lung were distributed exponentially, as shown in Figure 1. Nineteen to 21 nodules were added to the images obtained in each examination, and the locations of the nodules were equalized with regard to the following parameters: (a) level of the nodules, as defined by dividing the scanned images into an upper zone (above the aortic arch), middle zone, and lower zone (below the right inferior pulmonary vein); (b) central or peripheral distribution of the nodules, in which the peripheral area was defined as the outer 2 cm of lung parenchyma along the chest wall; and (c) ventral or dorsal distribution of the nodules, in which the border was defined as the center of the anteroposterior diameter of the thorax (Table 2). For every examination performed in the same volunteer, simulated nodules were placed at the same positions to permit the effects of tube current on the images to be evaluated free of any possible effects due to differences in the locations of the nodules (Figs 2, 3).
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Graph of CT numbers of simulated nodules. The simulated nodules are 180 and 66 HU higher in CT number than lung parenchyma at the center and edge of the nodules, respectively. The radius of each nodule is 5 pixels.
|
|
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2d. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2e. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2f. Volunteer 4. (a-c) Upper zone of the lung. (d-f) Middle zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 18, and (c, f) 10 mAs. The arrows in a and d indicate the simulated nodules. An increase in noise with a reduction in the effective tube current is more remarkable in a-c than in d-f. It is possible to detect all simulated nodules in the middle zone, even in f. In the upper zone, the nodules can hardly be recognized in c because of an increase in noise.
|
|
View larger version (96K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3e. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3f. Volunteer 3. (a-c) Middle zone of the lung. (d-f) Lower zone of the lung. The transverse CT scans were acquired at effective tube currents of (a, d) 50, (b, e) 20, and (c, f) 10 mAs. The arrow in a and the arrows in d indicate the simulated nodules. It is possible to detect all simulated nodules in the middle zone, even in c, while in f, some nodules in the lower zone are hardly detectable because of an increase in noise. In a, the nodular opacity (arrowhead) at the right inferior pulmonary vein is a vessel branch seen end-on.
|
|
Six experienced radiologists (S.A., T. Kodaira, T. Isomura, T. Kato, K.Y., K.M.) who were blinded to the scanning parameters used to acquire the CT images assessed the image quality independently by using a four-point scale, where 4 was excellent, 3 was good, 2 was adequate, and 1 was poor. They assessed the image quality with the understanding that the images were to be used in mass screening for lung cancer. The readers also reported the location of each suspected nodule and indicated their level of confidence that a nodule was actually present at that location by using a continuous rating scale from 0 to 1. This rating scale represented the reader's estimates of the probability that a nodule would be present, and 0 and 1 corresponded to the definite absence and definite presence, respectively, of a computer-generated nodule on the CT images.
To avoid causing a learning bias, we conducted the study in the following way. For the purpose of preventing the readers from learning where the nodules were located, the CT images obtained in the same volunteer were not evaluated successively but were randomly mixed with images from other volunteers, with no less than five other images interpreted. In addition, the reading order differed among the six readers to compensate for a learning bias in each reader.
On the basis of the results that another author (S.I.) obtained in the same study preliminarily, the reading time was limited to less than 8 minutes for each examination. This limitation allowed us to prevent the possibility that the degradation in the image quality due to a reduction in the tube current would be compensated by prolongation of the reading time.
In this study, we divided each CT section into eight compartments as follows: (a) right and left lungs, (b) ventral and dorsal, and (c) central and peripheral. We then evaluated each reader's judgment of the simulated nodules in each compartment with the Brier score (12,13), which was calculated by using the following equation:
where n is the number of compartments on each image, C is the confidence level for the presence of a nodule in each compartment, and Ti is 1 if a compartment contained a simulated nodule and 0 if a compartment did not contain a simulated nodule.
The lower this score, the better the reader's judgment of nodules. In this study, we collectively statistically analyzed all interpretations to clarify whether there was a significant difference for each item. The various effective tube currents were compared in terms of image quality rating and Brier score by using a repeated-measures, two-way analysis of variance. The influence of the effective tube current on the detectability of simulated nodules was also analyzed according to location. A P value of less than .05 was accepted as indicating a statistically significant difference.
One author (S.I.) measured the cross-sectional diameter in centimeters at the base of the lung and determined the SD of the CT number. The former was calculated as the square root of the sum of the square of the anteroposterior diameter and the square of the transverse diameter in each volunteer, according to the method by Haaga et al (14).
The SD of each CT number was determined by using a circular region of interest 7 pixels in diameter at the following levels: (a) the apex of the lung, defined as the first section containing lung parenchyma; (b) the branching of the aortic arch; (c) the bronchus intermedius; (d) the right inferior pulmonary vein; (e) the upper border of the right side of the diaphragm; and (f) the base of the lung, defined as the last section containing lung parenchyma. We selected homogeneous anatomic structures so that the SD of the CT number precisely reflected the image noise. Thus, the circular region of interest was placed over the trachea at the level of the apex of the lung and at the level of the branching of the aortic arch and over the descending aorta at the other four levels, with care taken to minimize partial volume artifacts.
The cross-sectional diameters and SDs were compared among various effective tube currents according to the location by using a repeated-measures, two-way analysis of variance.
|
RESULTS
|
|---|
The results for the assessment of image quality and the detection of simulated nodules are summarized in Table 3. A trend toward degradation in these two parameters was noted as the effective tube current was reduced. However, compared with those at 50 mAs, the assessment of image quality and the detection of simulated nodules were not significantly different until the effective tube current was reduced to below 20 mAs and to 18 mAs, respectively (Figs 4, 5).
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Assessment of image quality. Graph of image quality score versus the effective tube current. Compared with that at a tube current of 50 mAs, the image quality shows no significant difference until the effective tube current is reduced to below 20 mAs.
|
|
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Detectability of simulated nodules. Graph of Brier score versus the effective tube current. Compared with that at a tube current of 50 mAs, the detection of simulated nodules is not significantly different until the effective tube current is reduced to below 18 mAs.
|
|
With regard to the detection of simulated nodules, the minimum effective tube current with no significant degradation differed among the locations in the lung; this difference was marked among levels of the lung. At any effective tube current, the results were worse in the upper and lower zones of the lung than in the middle zone. Thus, while the detectability of nodules was not significantly degraded by reducing the effective tube current to 12 mAs in the middle zone, the minimum effective tube currents were 20 mAs and 18 mAs in the upper and lower zones, respectively (Fig 6).
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Detectability of simulated nodules among upper, middle, and lower zones of the lung. Graph of Brier score versus the effective tube current. The minimum effective tube currents that maintain the detectability of simulated nodules equivalent to that at 50 mAs are 20, 12, and 18 mAs in the upper, middle, and lower zones, respectively.
|
|
The minimum effective tube currents with significant degradation in the detection of nodules were 12 mAs and 10 mAs in the central and peripheral areas of the lung, respectively, in comparison with the detection of nodules at 50 mAs (Fig 7).
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7. Detectability of simulated nodules in central versus peripheral areas of the lung. Graph of Brier score versus the effective tube current. The minimum effective tube currents with no significant degradation in the detectability of simulated nodules are 18 and 12 mAs in the central and peripheral areas of the lung, respectively, in comparison with the detectability at 50 mAs.
|
|
By comparing the ventral and dorsal areas, the detectability of nodules was not significantly degraded by reducing the effective tube current to 18 mAs in either area. However, there was a tendency for the detectability in the dorsal area to be worse than that in the ventral area; this difference became more apparent as the tube current was reduced (Fig 8).
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Detectability of simulated nodules in ventral and dorsal areas of the lung. Graph of Brier score versus the effective tube current. The detectability of simulated nodules is not significantly degraded by reducing the effective tube current to 18 mAs in both the ventral and dorsal areas, in comparison with the detectability at 50 mAs.
|
|
There was no statistically significant difference among the various effective tube currents with regard to the cross-sectional diameter at the base of the lung (Table 1). The results for the the SDs of the CT numbers are summarized in Table 4 and Figure 9. A reduction in effective tube current resulted in an increase in the SD at all locations. At any effective tube current, the SDs at the apex and at the base of the lung were greater and more variable than the SDs at other locations, and the difference in SD between the (a) apex and base and (b) other locations became more apparent as the effective tube current was reduced. If an acceptable cutoff for noise in terms of image quality is considered to be an SD of about 100, the minimum tube current required for screening helical CT in the lung is about 50 mAs at the apex and base but only about 10 mAs at other locations.
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9. Graph of the SDs of the CT numbers (CT value) versus the effective tube current for the apex of the lung, the branching of the aortic arch, the bronchus intermedius, the right (rt) inferior pulmonary vein, the upper border of the right (rt) part of the diaphragm, and the base of the lung. The SDs at the apex and base of the lung are greater and more variable than those at other locations.
|
|
|
DISCUSSION
|
|---|
Findings of previous studies have demonstrated that mass screening for peripheral lung cancers by employing chest radiography is not sufficient to reduce the mortality rate and that more effective screening methods are required (2,15). Recently, lung cancer screening by employing helical CT has been attempted in Japan (3,4). Such screening helical CT is currently performed with scanning parameters of 120 kVp, 50 mAs, 1 second per rotation, 10-mm collimation, and a pitch of 2.0 at most institutions in Japan to reduce both the scanning time and the radiation dose.
For this screening method to be applied more generally, however, it is desirable to achieve further reductions in radiation dose. A reduction in radiation dose can be achieved by increasing the pitch and reducing the tube current. However, even if images are reconstructed with 180° linear interpolation, a further increase in pitch may result in an increase in the risk of missing small lung cancers that manifest as ground-glass opacity because of the broadening of the section-sensitivity profile. Therefore, it is necessary to reduce the tube current in helical CT for lung cancer screening.
A reduction in tube current results in the degradation of image quality because of an increase in noise; thus, we evaluated images obtained in various low-dose CT examinations to investigate the minimum tube current required for helical CT in lung cancer screening. Haaga et al (14) reported that there appears to be a linear relationship between noise and body size. Therefore, we elected to examine healthy volunteers several times at various tube currents to assess the effects of reducing the tube current on CT images, without any possible effects of differences among individual subjects.
To create the experimental stimuli, simulated nodules were superimposed electronically onto sets of normal lung CT images (16,17). This method allowed us to manipulate the characteristics of the nodules, such as the location, attenuation, and size, in consideration of the acceptable image quality required for helical CT in lung cancer screening and to assess the influence of increasing noise on diagnostic capabilities more objectively than with the simple assessment of image quality.
Noguchi et al (18) reported that adenocarcinoma of the lung with little fibrosis has been curable. These cancers appear as focal areas of ground-glass opacity on CT images (19), and the CT numbers of these opacities have been reported to be at least 200 HU higher than the CT number of lung parenchyma (20). Therefore, we defined the increase in CT number for simulated nodules as at most 180 HU to ensure the image quality was that required for the detection of lesions that manifest as ground-glass opacity. Because the objective in helical CT for lung cancer screening is to detect cancers 10 mm in diameter, simulated nodules of about 6 mm in diameter were added to the images.
In this study, no significant degradation in the assessment of image quality or the detection of simulated nodules was seen when the effective tube current was reduced to 20 mAs and to 18 mAs, respectively, in comparison with a tube current of 50 mAs. This finding suggests that it is possible to achieve further reductions in radiation dose in helical CT for lung cancer screening, which is currently performed at a tube current of 50 mAs.
The results of the present study, however, showed that, compared with the middle zone, the upper and lower zones of the lung require a higher radiation dose to maintain a detectability of the simulated nodules equivalent to that at 50 mAs. Furthermore, the SDs of CT numbers at the apex and base of the lung were larger than those at other locations, and the differences were more apparent when the tube current was reduced. These results are thought to be due to the anatomic characteristics of the lung. Specifically, since image noise is increased by the shoulder at the apex of the lung and by the liver and the heart at the base of the lung, a higher dose is required to achieve an image quality equivalent to that at other locations of the lung. Therefore, to minimize the radiation dose, it is necessary to change the tube current when examining different locations in the lung with low-dose helical CT for lung cancer screening.
The radiation dose required to maintain the detectability of nodules equivalent to that at 50 mAs was higher in the central area than in the peripheral area. This result suggests that the influence of increasing noise on diagnostic capabilities is more remarkable in the central area because the anatomic structures of the area are more complex. The main reason that the detectability of nodules was worse in the dorsal area than in the ventral area was the increase in noise caused by the vertebrae in the dorsal area. These phenomena are more pronounced on low-dose CT scans because the noise value reaches the limit for acceptable image quality.
Findings of previous studies have demonstrated that the image quality at helical CT performed at a tube current of 50 mAs is certainly adequate for the detection of curable peripheral lung cancers (3,4,10,11). From the results of the present study and a review of the literature (5,6,21), we believe that it is possible to reduce the tube current to below 50 mAs in helical CT for lung cancer screening. It will be necessary to determine the limit of low-dose helical CT in the clinical setting to ensure that curable lung cancers are not missed. Then, from the results of this study, we must consider the possibility that the image quality of helical CT performed at a tube current of, for example, 20 mAs is acceptable for lung cancer screening at most levels of the lung but not at the apex and base. To overcome this limitation, helical CT scanners must be provided with a system for changing the tube current during scanning of the lung. Furthermore, the technique described here should also prove to be useful for reducing the radiation dose of chest CT in standard clinical practice while maintaining diagnostic accuracy.
One reason the detection of simulated nodules is worst in the upper zone of the lung is the presence of partial volume artifacts due to the thoracic anatomy. Such artifacts are aggravated by the relatively large pitch of 2.0 used with this screening method. Recently developed subsecond rotation, helical CT scanners may make it possible to use a smaller pitch, such as 1.5, without prolonging the scanning time, although the noise would increase because of the reduction in dose per rotation at the same tube current. We are currently performing further study to determine whether the benefit of reduced partial volume artifacts outweighs the disadvantage of increased noise in subsecond rotation helical CT.
Because the purpose of this study was to investigate the minimum tube current required for helical CT in lung cancer screening, we used the same scanning parameters, except for the tube current, employed in screening helical CT at most institutions in Japan. In addition, in the CT scanner used in the present study, it is difficult to scan the entire lung during a single breath hold by using 5-mm collimation. However, since findings of a previous study (22) have shown that 4- or 5-mm collimation and 4- or 5-mm reconstruction intervals will increase the detection rates of pulmonary nodules and that the multidetector CT scanners may make it possible to use the collimation of less than 4 mm and the higher pitch, the ideal parameters for helical CT in lung cancer screening will be reevaluated with the development of the CT scanners. Nevertheless, the results of this study have application in determining new scanning parameters.
In conclusion, the minimum tube current required for helical CT in lung cancer screening differs with the location in the lung, and an ideal CT protocol for the lung requires a method for changing the tube current during helical scanning.
|
Footnotes
|
|---|
Author contributions: Guarantor of integrity of entire study, S.I.; study concepts, S.I.; study design, S.I., M.I.; definition of intellectual content, S.I., M.I.; literature research, S.I., M.I.; clinical studies, S.I., M.I., S.A., T. Kodaira, T. Isomura, T. Kato, K.Y., K.M.; data acquisition, S.I.; data analysis, S.I., M.I.; statistical analysis, M.I.; manuscript preparation and editing, S.I.; manuscript review, S.I., M.I., T. Ishigaki.
|
References
|
|---|
-
Kurokawa T, Matsuno Y, Noguchi M, Mizuno S, Shimosato Y. Surgically curable "early" adenocarcinoma in the periphery of the lung. Am J Surg Pathol 1994; 18:431-438.[Medline]
-
Soda H, Tomita H, Kohno S, Oka M. Limitation of annual screening chest radiography for the diagnosis of lung cancer. Cancer 1993; 72:2341-2346.[Medline]
-
Matsumoto M, Horikoshi H, Moteki T, et al. A pilot study with lung-cancer screening CT (LSCT) at the secondary screening for lung cancer detection. Nippon Acta Radiol 1995; 55:172-179[Japanese].
-
Kaneko M, Eguchi K, Ohmatsu H, et al. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 1996; 201:798-802.[Abstract/Free Full Text]
-
Naidich DP, Marshall CH, Gribbin C, Arams R, McCauley DI. Low-dose CT of the lungs: preliminary observation. Radiology 1990; 175:729-731.[Abstract/Free Full Text]
-
Zwirewich CV, Mayo JR, Müller NL. Low-dose high-resolution CT of lung parenchyma. Radiology 1991; 180:413-417.[Abstract/Free Full Text]
-
Mayo JR, Hartman TE, Lee KS, Primack SL, Vedal S, Müller NL. CT of the chest: minimal tube current required for good image quality with the least radiation dose. AJR Am J Roentgenol 1995; 164:603-607.[Abstract/Free Full Text]
-
Kato R, Katada K, Anno H, Suzuki S, Ida Y, Koga S. Radiation dosimetry at CT fluoroscopy: physician's hand dose and development of needle holders. Radiology 1996; 201:576-578.[Abstract/Free Full Text]
-
Itoh S, Koyama S, Tusaka M, Maekoshi H, Satake H, Ishigaki T. Helical CT for lung-cancer screening: third report—fundamental study for ultra-low-dose CT by application of small tube current and filter. Nippon Acta Radiol 1996; 56:961-966[Japanese].
-
Mori K. Detection of pulmonary nodules using helical CT. Jpn J Lung Cancer 1995; 35:149-156[Japanese].
-
Itoh S, Ikeda M, Isomura T, et al. Screening helical CT for mass screening of lung cancer: application of low-dose and single-breath-hold scanning. Radiat Med 1998; 16:75-83.[Medline]
-
Spiegelhalter DJ. Probabilistic prediction in patient management and clinical trials. Stat Med 1986; 5:421-433.[Medline]
-
Redelmeier DA, Bloch DA, Hickam DH. Assessing predictive accuracy: how to compare Brier score. J Clin Epidemiol 1991; 44:1141-1146.[Medline]
-
Haaga JR, Miraldi F, MacIntyre W, LiPuma JP, Bryan PJ, Wiesen E. The effect of mAs variation upon computed tomography image quality as evaluated by in vivo and in vitro studies. Radiology 1981; 138:449-454.[Abstract/Free Full Text]
-
Fontana RS, Sanderson DR, Woolner LB, et al. Screening for lung cancer: a critique of the Mayo lung project. Cancer 1991; 67:1155-1164.[Medline]
-
Naidich DP, Rusinek H, McGuinness G, Leitman B, McCauley DI, Henschke CI. Variable affecting pulmonary nodule detection with computed tomography: evaluation with three-dimensional computer simulation. J Thorac Imaging 1993; 8:291-299.[Medline]
-
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]
-
Noguchi M, Morikawa A, Kawasaki M, et al. Small adenocarcinoma of the lung: histologic characteristics and prognosis. Cancer 1995; 75:2844-2852.[Medline]
-
Jang HJ, Lee KS, Kwon OJ, Rhee CH, Shim YM, Han J. Bronchioloalveolar carcinoma: focal area of ground-glass attenuation at thin-section CT as an early sign. Radiology 1996; 199:485-488.[Abstract/Free Full Text]
-
Nagashima S, Eguchi K, Hirano H, Kaneko M, Kondo H, Noguchi M. Measurement of the CT values in adenocarcinoma of the lung showing ground-glass attenuation at thin-section CT (abstr). Jpn J Lung Cancer 1995; 35:661[Japanese].
-
Rusinek H, Naidich DP, McGuinness G, et al. Pulmonary nodule detection: low-dose versus conventional CT. Radiology 1998; 209:243-249.[Abstract/Free Full Text]
-
Buckley JA, Scott Jr WW, Siegelman SS, et al. Pulmonary nodules: effect of increased data sampling on detection with spiral CT and confidence in diagnosis. Radiology 1995; 196:395-400.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
T. Kubo, P.-J. P. Lin, W. Stiller, M. Takahashi, H.-U. Kauczor, Y. Ohno, and H. Hatabu
Radiation Dose Reduction in Chest CT: A Review
Am. J. Roentgenol.,
February 1, 2008;
190(2):
335 - 343.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. A. Gietema, A. M. Schilham, B. van Ginneken, R. J. van Klaveren, J. W. J. Lammers, and M. Prokop
Monitoring of Smoking-induced Emphysema with CT in a Lung Cancer Screening Setting: Detection of Real Increase in Extent of Emphysema
Radiology,
September 1, 2007;
244(3):
890 - 897.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Barentsz, S. Takahashi, W. Oyen, R. Mus, P. De Mulder, R. Reznek, M. Oudkerk, and W. Mali
Commonly Used Imaging Techniques for Diagnosis and Staging
J. Clin. Oncol.,
July 10, 2006;
24(20):
3234 - 3244.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. H. Mulkens, P. Bellinck, M. Baeyaert, D. Ghysen, X. Van Dijck, E. Mussen, C. Venstermans, and J.-L. Termote
Use of an Automatic Exposure Control Mechanism for Dose Optimization in Multi-Detector Row CT Examinations: Clinical Evaluation
Radiology,
October 1, 2005;
237(1):
213 - 223.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Remy-Jardin, A. Sobaszek, A. Duhamel, I. Mastora, C. Zanetti, and J. Remy
Asbestos-related Pleuropulmonary Diseases: Evaluation with Low-Dose Four-Detector Row Spiral CT
Radiology,
October 1, 2004;
233(1):
182 - 190.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
I. Mastora, M. Remy-Jardin, V. Delannoy, A. Duhamel, C. Scherf, C. Suess, and J. Remy
Multi-Detector Row Spiral CT Angiography of the Thoracic Outlet: Dose Reduction with Anatomically Adapted Online Tube Current Modulation and Preset Dose Savings
Radiology,
January 1, 2004;
230(1):
116 - 124.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. P. Ko, H. Rusinek, E. L. Jacobs, J. S. Babb, M. Betke, G. McGuinness, and D. P. Naidich
Small Pulmonary Nodules: Volume Measurement at Chest CT--Phantom Study
Radiology,
September 1, 2003;
228(3):
864 - 870.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Tack, V. De Maertelaer, and P. A. Gevenois
Dose Reduction in Multidetector CT Using Attenuation-Based Online Tube Current Modulation
Am. J. Roentgenol.,
August 1, 2003;
181(2):
331 - 334.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. R. Mayo, J. Aldrich, and N. L. Muller
Radiation Exposure at Chest CT: A Statement of the Fleischner Society
Radiology,
July 1, 2003;
228(1):
15 - 21.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Itoh, M. Ikeda, Y. Mori, K. Suzuki, A. Sawaki, S. Iwano, H. Satake, S. Arahata, T. Isomura, M. Ozaki, et al.
Lung: Feasibility of a Method for Changing Tube Current during Low-Dose Helical CT
Radiology,
September 1, 2002;
224(3):
905 - 912.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. G. Ravenel, E. M. Scalzetti, W. Huda, and W. Garrisi
Radiation Exposure and Image Quality in Chest CT Examinations
Am. J. Roentgenol.,
August 1, 2001;
177(2):
279 - 284.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. R. Haaga
Radiation Dose Management: Weighing Risk Versus Benefit
Am. J. Roentgenol.,
August 1, 2001;
177(2):
289 - 291.
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
