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


     

Published online before print April 19, 2007, 10.1148/radiol.2433061165
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2433061165v1
243/3/775    most recent
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 McCollough, C. H.
Right arrow Articles by Flohr, T. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCollough, C. H.
Right arrow Articles by Flohr, T. G.
(Radiology 2007;243:775-784.)
© RSNA, 2007

Medical Physics

Dose Performance of a 64-Channel Dual-Source CT Scanner1

Cynthia H. McCollough, PhD, Andrew N. Primak, PhD, Osama Saba, PhD, Herbert Bruder, PhD, Karl Stierstorfer, PhD, Rainer Raupach, PhD, Christoph Suess, PhD, Bernhard Schmidt, PhD, Bernd M. Ohnesorge, PhD, and Thomas G. Flohr, PhD

1 From the CT Clinical Innovation Center, Department of Radiology, Mayo Clinic College of Medicine, 200 First St SW, Rochester, MN 55905 (C.H.M., A.N.P.); Siemens Medical Solutions, Malvern, Pa (O.S.); Siemens Medical Solutions, Forchheim, Germany (H.B., K.S., R.R., C.S., B.S., B.M.O., T.G.F.); and Department of Diagnostic Radiology, Eberhard-Karls-Universität Tübingen, Tübingen, Germany (T.G.F.). From the 2005 RSNA Annual Meeting. Received July 6, 2006; revision requested August 31; revision received October 9; accepted November 2; final version accepted December 4. Address correspondence to C.H.M. (e-mail: mccollough.cynthia{at}mayo.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 IMPLICATION FOR PATIENT CARE
 References
 
Purpose: To prospectively compare the dose performance of a 64-channel multi–detector row computed tomographic (CT) scanner and a 64-channel dual-source CT scanner from the same manufacturer.

Materials and Methods: To minimize dose in the cardiac (dual-source) mode, the evaluated dual-source CT system uses a cardiac beam-shaping filter, three-dimensional adaptive noise reduction, heart rate–dependent pitch, and electrocardiographically based modulation of the tube current. Weighted CT dose index per 100 mAs was measured for the head, body, and cardiac beam-shaping filters. Kerma-length product was measured in the spiral cardiac mode at four pitch values and three electrocardiographic modulation temporal windows. Noise was measured in an anthropomorphic phantom. Data were compared with data from a 64-channel multi–detector row CT scanner.

Results: For the multi–detector row and dual-source CT systems, respectively, weighted CT dose index per 100 mAs was 14.2 and 12.2 mGy (head CT), 6.8 and 6.4 mGy (body CT), and 6.8 and 5.3 mGy (cardiac CT). In the spiral cardiac mode (no electrocardiographically based tube current modulation, 0.2 pitch), equivalent noise occurred at volume CT dose index values of 23.7 and 35.0 mGy (coronary artery calcium CT) and 58.9 and 61.2 mGy (coronary CT angiography) for multi–detector row CT and dual-source CT, respectively. The use of heart rate–dependent pitch values reduced volume CT dose index to 46.2 mGy (0.265 pitch), 34.0 mGy (0.36 pitch), and 26.6 mGy (0.46 pitch) compared with 61.2 mGy for 0.2 pitch. The use of electrocardiographically based tube current–modulation and temporal windows of 110, 210, and 310 msec further reduced volume CT dose index to 9.1–25.1 mGy, dependent on the heart rate.

Conclusion: For electrocardiographically gated coronary CT angiography, image noise equivalent to that of multi–detector row CT can be achieved with dual-source CT at doses comparable to or up to a factor of two lower than the doses at multi–detector row CT, depending on heart rate of the patient.

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 IMPLICATION FOR PATIENT CARE
 References
 
Electrocardiographically gated cardiac computed tomographic (CT) examinations with multi–detector row CT systems were introduced in 1999 (1–3). Despite promising initial results, challenges with respect to motion artifacts at higher heart rates remained. In addition, concerns were raised with respect to the radiation dose levels from cardiac CT examinations (4,5) (Table 1).


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

  		Table 1. Typical Effective Dose Values for Noncardiac and Cardiac CT Examinations

 
Multi–detector row 16-channel CT systems with gantry rotation times as low as 0.375 second led to improved spatial and temporal resolution compared with the spatial and temporal resolution of earlier scanners and resulted in reduced examination times (8,9). The latest generation of 64-channel multi–detector row CT systems led to further increased spatial resolution (0.4-mm isotropic voxels with the use of advanced z-sampling techniques [10]) and improved temporal resolution caused by gantry rotation times as low as 0.33 second. However, these decreases in gantry rotation time necessitated decreases in pitch to avoid gaps in the volume coverage at lower heart rates (11,12), and the reduction in pitch led to an increase in the radiation dose for 64-channel systems relative to 16-channel systems. Although image quality and clinical robustness seem to be significantly improved with 64-channel CT systems compared with previous generations of multi–detector row CT systems, several authors still propose the administration of beta-blockers for cardiac CT (13–15).

To eliminate the need to reduce heart rate prior to cardiac CT, a temporal resolution of less than 100 msec at all heart rates is desired. Although multisegment reconstruction techniques (12,16) can help in the achievement of such temporal resolutions, they require an absolutely steady heart rate. Thus, increased gantry rotation speed appears to be a preferable approach for robust clinical performance (17). However, rotation times of less than 0.2 second are required to provide temporal resolution of less than 100 msec with single-segment reconstruction. These rotation times (<0.2 second) create mechanical forces more than 75 times the force of gravity and appear to be beyond today's mechanical limits.

An alternative concept to improve temporal resolution is the use of multiple x-ray sources and detectors (18,19). Recently, a dual-source CT system (Somatom Definition; Siemens Medical Solutions, Forchheim, Germany) has been introduced (20). The purpose of our study was to prospectively compare the dose performance of a 64-channel multi–detector row CT scanner and a 64-channel dual-source CT scanner from the same manufacturer.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 IMPLICATION FOR PATIENT CARE
 References
 
The dual-source CT system evaluated in this study was provided by Siemens Medical Solutions (Forchheim, Germany) as part of an ongoing research collaboration within the Department of Radiology, Mayo Clinic College of Medicine, Rochester, Minn. At all times, control of the data and information for publication remained with those authors who are not employees of Siemens Medical Solutions (C.H.M., A.N.P.). These two authors receive partial research funding from Siemens Medical Solutions.

Dual-Source CT System
The evaluated dual-source CT system was equipped with two x-ray tubes and two corresponding detectors. The two acquisition systems were mounted onto the rotating gantry with an angular offset of 90° (Fig 1). One detector array (corresponding to tube A) covered the entire field of view (FOV) of the scan (50 cm), while the other detector array (corresponding to tube B) was restricted to a smaller central FOV (26 cm) because of space limitations on the gantry. Each detector contained 40 detector rows; the 32 central rows were 0.6 mm wide and the four outer rows on both sides were 1.2 mm wide. By using the z-flying focal spot technique (8), two overlapping sets of 32 x 0.6-mm measurements were combined to create 64 projections along the z-axis, per gantry rotation, with an overlap of 0.3 mm for each 0.6-mm measurement. The shortest gantry rotation time was 0.33 second. Two generators provided up to 80-kW peak power to each of the rotating-envelope x-ray tubes (Straton; Siemens Medical Solutions) (21). When only tube A was operated, the performance of the system was essentially the same as that previously described for the 64-channel single-source CT scanner from the same manufacturer (10).


Figure 1
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Schematic illustration of the acquisition principle of dual-source CT by using two tubes (A and B) and two corresponding detectors. Because of the 90° offset between the two detectors, the half-scan sinogram in parallel geometry can be split up into two quarter-scan sinograms (indicated as black and gray quarter circles), which are simultaneously acquired by the two acquisition systems.

 
In general, partial scans (also called half scans) are used for electrocardiographically gated CT image reconstruction, and the acquisition time window of the data that contribute to the reconstruction determines the temporal resolution ({Delta}Tima). If redundant data are neglected, temporal resolution {Delta}Tima at the center of rotation can be as good as half of the gantry rotation time, or trot/2, for a single-source CT scanner. For a 0.33-second rotation, {Delta}Tima = trot/2 = 165 msec. With a dual-source CT scanner, the half-scan sinogram can be split into two quarter-scan sinograms (Fig 1). These two quarter-scan sinograms are simultaneously acquired with the two detectors and are joined together by means of a smooth transition function to prevent artifacts from potential discontinuities at the respective start and end projections. Because detector B does not cover the entire 50-cm FOV of the scan, its projections are potentially truncated and have to be supplemented by using data acquired with detector A for the same projection angle (ie, a quarter rotation earlier). With this approach, temporal resolution {Delta}Tima equivalent to a quarter of the gantry rotation time, or trot/4, is achieved for the FOV covered by both detectors. For trot = 0.33 second, this temporal resolution is {Delta}Tima = trot/4 = 83 msec. Because data from only one cardiac cycle are required to reconstruct an image (ie, single-segment reconstruction), the temporal resolution does not depend on heart rate.

The tube current–time product shown on the user interface is directly proportional to dose. It is calculated as the tube current multiplied by the time (in seconds per gantry rotation) during which the x-ray tube is energized. For cardiac dual-source CT scans, the tube current–time product per rotation sums the tube current–time product from tube A and tube B and is proportional to dose. The tube current–time product per image, however, is related to image noise and is determined by the number of projections used for image formation. Table 2 summarizes the angular range of projection data used from tube A (fan angle, 52°) and tube B (fan angle, 27°) in cardiac dual-source CT.


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

  		Table 2. Comparison of the Tube Current–Time Product per Rotation and the Tube Current–Time Product per Image between Noncardiac Multi–Detector Row CT and Cardiac Multi–Detector Row CT and Dual-Source CT

 
Dose Reduction Methods
Targeted FOV cardiac beam shaping.—Because the thickness of the patient is less at the patient periphery relative to the center of the patient, the x-ray beam is attenuated (ie, "shaped") prior to reaching the patient to eliminate unnecessary radiation to the periphery of the body (Fig 2). In cardiac CT, the region of interest is centered within the thorax, and radiation can be further restricted to the cardiac FOV. Thus, the radiation dose beyond the cardiac FOV can be reduced by using a more aggressive beam-shaping filter.


Figure 2
View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Body and targeted field-of-view (cardiac) beam-shaping filters. The intensity of the x-ray beam is indicated with the gray-scale gradient, with darker shading indicating higher x-ray beam intensity.

 
To determine the degree of dose reduction from the cardiac beam-shaping filter, the weighted CT dose index in milligrays was measured according to publication 60601-2-44 (CT safety standard) of the International Electrotechnical Commission (22). Weighted CT dose index is also described elsewhere (4,5,22–24). Conventional CT dose metrics are fully applicable to dual-source CT because the doses from each tube add together in a straightforward linear manner, with each dose independent of the other. Hence, measurements with each tube separately energized or with both tubes simultaneously energized result in the same combined dose values. The standard 16-cm- and 32-cm-diameter acrylic CT dose index phantoms were used to measure weighted CT dose index in the head and body acquisition modes, respectively. The 32-cm-diameter phantom was used for cardiac dose measurements. A 10-cm-long pencil ionization chamber and electrometer (Cap CII-4P-9348; PTW, Freiburg, Germany) were used to record the air kerma–length product measured in milligray-centimeters. The CT dose index integrated over 100 mm was calculated by dividing the air kerma–length product by the total nominal width of the x-ray beam. The data for CT dose index integrated over 100 mm at the center and 12-o'clock positions within the FOV were acquired in the sequential acquisition mode with no table translation and were recorded independently by three authors (C.H.M., with 16 years of experience in the evaluation of CT equipment; A.N.P., with 3 years of similar experience; K.S., with 12 years of similar experience). Data were acquired at a tube voltage of 120 kVp, a tube current of 400 mA, and a gantry rotation time of 1 second.

Three-dimensional adaptive noise reduction algorithm.—A three-dimensional adaptive noise reduction algorithm was commercially implemented in the evaluated dual-source CT system (in the B26 reconstruction kernel, where B26 is the name given to the kernel by the manufacturer and is shown on the operator's console). In this algorithm, linear variances are calculated for numerous directions in the three-dimensional image space to determine the orientation of edges. The minimum variance is assumed to be oriented tangentially to the contour with highest contrast. Different reconstruction filters are automatically used to generate three intermediate data sets that are not shown to the user. Depending on the local distribution of variances, intermediate data are mixed by the algorithm by using local weighting factors to obtain the final reconstructed image; the optimal adapted filter is the combination that maximizes the noise reduction with negligible deterioration of signal intensity (25). The preservation of image fidelity was previously demonstrated by using clinical image data (25). The B25 kernel of the 64-channel multi–detector row CT system that we evaluated uses a similar algorithm that is implemented in only two dimensions (within the image plane [x,y] but not along the z-axis).

To determine the dose reduction achieved with the three-dimensional algorithm, the dose required for a given noise level was determined with the B25 (multi–detector row CT) and B26 (dual-source CT) kernels. The noise versus the tube current–time product was measured (A.N.P., C.H.M., B.S., with 9 years of experience in CT equipment evaluation) for dual-source CT and 64-channel multi–detector row CT (Sensation 64; Siemens Medical Solutions) systems from the same manufacturer. Measurements were performed with an anthropomorphic cardiac CT phantom (Cardio; QRM, Möhrendorf, Germany) (Fig 3a), which simulates the thorax of a medium-sized adult man (26). The phantom was scanned by using electrocardiographically gated spiral acquisitions only at a pitch of 0.2, as noise in cardiac spiral CT is independent of pitch (11), and without tube current modulation. (Pitch is defined as the ratio of the table increment per tube rotation to total nominal beam width, in accordance with data in International Electrotechnical Commission publication 60601-2-44 [22].) Noise was measured as the standard deviation of the pixel values within a water-equivalent cylinder embedded in the central portion of the cardiac phantom (Fig 3b) by using a 1.65-cm2 region of interest. Data were acquired at multiple tube current–time product values spanning 25–200 mAs per tube per rotation for the coronary artery calcium CT (13 data sets for multi–detector row CT, eight data sets for dual-source CT) and the CT angiographic (nine data sets for multi–detector row CT, eight data sets for dual-source CT) examination protocols.


Figure 3A
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: (a) Anthropomorphic cardiac phantom and (b) transverse CT image show the water-equivalent cylindric insert (arrow) used for noise versus tube current–time product measurements.

 

Figure 3B
View larger version (52K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: (a) Anthropomorphic cardiac phantom and (b) transverse CT image show the water-equivalent cylindric insert (arrow) used for noise versus tube current–time product measurements.

 
The tube current–time product setting used clinically at our institution for coronary artery calcium CT and CT angiographic examinations with the multi–detector row CT system was used to determine the corresponding image noise. The tube current–time product setting required to achieve the same image noise with the dual-source CT system was then determined for the coronary artery calcium CT and CT angiographic protocols. The doses for the two systems were then compared for equivalent image noise and spatial resolution.

Heart rate–dependent pitch values.—In cardiac CT, temporal resolution strongly depends on heart rate (8,12,16), with optimal temporal resolution achieved when the patient's heart rate and the gantry rotation time are properly desynchronized. For a multi–detector row CT system and a dual-source CT system (with both having a 0.33-second gantry rotation time) (20), the multi–detector row CT system reaches 83-msec temporal resolution by using dual-segment reconstruction only at heart rates of 66, 81, and 104 beats per minute, whereas the dual-source CT system provides 83-msec temporal resolution at all heart rates by using data from only one cardiac cycle (ie, single-segment reconstruction) (Fig 4).


Figure 4
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4: Temporal resolution as a function of the patient's heart rate for multi–detector row CT (dotted line) and dual-source CT (solid line) at 0.33-second gantry rotation time. The image reconstruction mode (ACV [Adaptive Cardio Volume; Siemens Medical Solutions]) employed with the evaluated multi–detector row CT system automatically switches from single-segment to dual-segment reconstruction for heart rates greater than 65 beats per minute (bpm), whereas dual-source CT always uses single-segment reconstruction.

 
In addition, for single-segment cardiac CT image reconstruction, the table feed (and its correlate of pitch) can be adapted to the heart rate, with an increase in pitch for higher heart rates (Fig 5) (1). This adaptation directly leads to a decrease in patient dose, as the average dose within the scan volume is directly proportional to 1/pitch. However, in order to allow for downward variations in heart rate during scan acquisition, or the option of multisegment image reconstruction to improve temporal resolution, most commercial systems keep pitch relatively low even when the heart rate is increased (eg, more than 65 beats per minute). That is, with single-source CT, pitch is not increased dramatically at higher heart rates because multisegment image reconstruction must be used to improve temporal resolution and avoid motion artifacts. Because maintaining a low pitch value is not necessary for dual-source CT, pitch could be increased from 0.2 to 0.46 for the evaluated dual-source CT system as the heart rate increased from 45 to 100 beats per minute.


Figure 5A
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Graphical description of an electrocardiographically gated spiral scan for high heart rates (>65 beats per minute) with (a) multi–detector row CT and (b) dual-source CT. The slope of the dotted lines (detector trajectories) is proportional to pitch and is limited by the need for continuous heart volume coverage. Note the significant slope (pitch) difference caused by the need to use dual-segment reconstruction at higher heart rates with multi–detector row CT versus the ability to use a single-segment dual-source CT reconstruction at high heart rates with dual-source CT. ECG = electrocardiographic signal, R = R wave of the electrocardiographic signal, R delay = time delay from detected R wave to the reconstruction window, Recon = reconstruction window.

 

Figure 5B
View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Graphical description of an electrocardiographically gated spiral scan for high heart rates (>65 beats per minute) with (a) multi–detector row CT and (b) dual-source CT. The slope of the dotted lines (detector trajectories) is proportional to pitch and is limited by the need for continuous heart volume coverage. Note the significant slope (pitch) difference caused by the need to use dual-segment reconstruction at higher heart rates with multi–detector row CT versus the ability to use a single-segment dual-source CT reconstruction at high heart rates with dual-source CT. ECG = electrocardiographic signal, R = R wave of the electrocardiographic signal, R delay = time delay from detected R wave to the reconstruction window, Recon = reconstruction window.

 
To confirm the expected dose decrease with increased pitch, air kerma–length product was measured in the spiral acquisition mode to directly assess the effects of pitch variation (A.N.P., C.H.M., O.S., with 5 years of experience in the evaluation of CT equipment; K.S., T.G.F., with 16 years of similar experience). The same ionization chamber, electrometer, and CT dose index body phantom were used as in the weighted CT dose index measurements. The spiral scan was initiated 7.5 cm outside the acrylic phantom and was terminated 7.5 cm beyond the opposite end of the phantom, such that the entire phantom and ionization chamber were irradiated. This scan volume was consistent for all measurements, and the air kerma–length product (dose) reduction caused by pitch variation was determined relative to the dose at a pitch of 0.2 (Table 3). No tube current modulation was used for these measurements. Electrocardiographic signals were provided by an electrocardiographic simulator, and data were collected for the cardiac mode of the dual-source CT system at 120 kV and 500 mA for each x-ray tube, 0.33-second gantry rotation, and 32 x 0.6-mm collimation with the z-flying focal spot technique (10). A total of five data sets were collected (one set for multi–detector row CT at a pitch of 0.2; four sets for dual-source CT at pitch values of 0.2, 0.265, 0.36, and 0.46).


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

  		Table 3. Heart Rate–dependent Pitch and Table Feed Settings for Dual-Source CT Scanner

 
Electrocardiographic modulation.—Modulation of the tube current throughout the cardiac cycle has been shown to help in the reduction of dose in cardiac CT by as much as 50% (27). As implemented in the evaluated dual-source CT scanner, the user can prescribe the width of the temporal window at which the maximum tube current is used (Fig 6) and can select a temporal window width as narrow as 110 msec.


Figure 6
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6: Depiction of electrocardiographic modulation of the tube current for the dual-source CT system. The width of the temporal window that has the maximum tube current (max mA) can be selected by the user, whereas the temporal width of the image reconstruction window is 83 msec. For full-quality images for coronary artery CT angiography, the reconstruction window (Recon) (black bars) should be within the maximum tube current window (gray bars). Min mA = minimum tube current, R = detected R wave of electrocardiographic signal.

 
The patient dose when using tube current modulation is proportional to the tube current averaged over the duty cycle, which is defined with the following calculation: [(tmin · mAmin) + (tmax · mAmax)]/(tmin + tmax), where tmin and tmax are the times at minimum tube current (mAmin) and maximum tube current (mAmax), respectively, within one cardiac cycle. The transition from minimum tube current to maximum tube current, and vice versa, is not instantaneous, so theoretical predictions that are based on this calculation will differ somewhat from experimental measurements because of the time required to ramp up and ramp down the tube current. To directly measure the dose reduction achieved from tube current modulation, the air kerma–length product was measured (A.N.P., C.H.M., O.S., K.S., T.G.F.) in the spiral acquisition mode as the temporal window width for the electrocardiographically based tube current modulation was varied from no modulation to 310-, 210-, and 110-msec window widths. The same methods and scan parameters used to measure the dose as a function of pitch were used for this evaluation. A total of 16 data sets were collected (four for multi–detector row CT at a pitch of 0.2 with a 400-msec temporal window and four heart rates; 12 for dual-source CT at four pitch values of 0.2, 0.265, 0.36, and 0.46 corresponding to the same four heart rates and 310-, 210-, and 110-msec temporal windows).

The generators and rotating envelope tubes (21) used in the evaluated dual-source CT system allow faster tube current transitions relative to the implementation in the evaluated multi–detector row CT system and in other multi–detector row CT systems with conventional x-ray tube designs. Figure 7 shows the differences between the evaluated dual-source CT system and a multi–detector row CT system that uses a nonrotating envelope x-ray tube (Akron; Siemens Medical Solutions). Use of a faster transition time decreases the total dose delivered to obtain the maximum tube current for the desired temporal window, as less time is spent ramping up to and down from the target tube current.


Figure 7
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7: Graph shows comparison of the ramp-up time between the multi–detector row CT (MDCT) and dual-source CT (DSCT) systems and plots the relative tube current as a function of time.

 
A limitation of electrocardiographically based modulation techniques occurs when patient heart rate is not stable, because then the optimal reconstruction window might occur during the reduced tube current window. The dual-source CT system addresses this concern by automatically increasing the tube current to the target level when a statistical trend-analysis algorithm recognizes an R-R interval that is significantly different from the previous rhythm.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 IMPLICATION FOR PATIENT CARE
 References
 
With use of the standard body beam-shaping filter, which was the same for the two systems, the weighted CT dose index was 6.8 mGy per 100 mAs for multi–detector row CT and 12.8 mGy per 100 mAs per tube for dual-source CT (100 mAs for each tube, for a total of 200 mAs). Thus, in spiral modes when both tubes of the dual-source CT system were simultaneously energized, the dose was a factor of 1.88 higher relative to that of multi–detector row CT (Table 4).


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

  		Table 4. Weighted CT Dose Index Measurements per 100 mAs per Tube for Multi–Detector Row CT and Dual-Source CT Systems

 
Targeted FOV Cardiac Beam Shaping
With the dual-source CT system, the addition of a cardiac beam-shaping filter was shown to reduce the dose by 17% relative to the body beam-shaping filter only, and the weighted CT dose index was reduced to 10.6 mGy per 100 mAs per tube in the cardiac mode (tube A plus tube B). Hence, use of the cardiac beam-shaping filter reduces the dose increase of dual-source CT relative to multi–detector row CT from a factor of 1.88 to a factor of 1.55. The reduced scan FOV of tube B provides some dose reduction relative to tube A, even without use of the cardiac filter, because of the restricted dimension of the x-ray beam in the scan plane.

Three-dimensional Adaptive Noise Reduction
When the same image reconstruction kernel (B35) was used, the measured noise was found to be essentially the same for dual-source CT as it was for multi–detector row CT (with use of the same tube current–time product setting for each dual-source CT tube as is used for the single tube with multi–detector row CT) (Fig 8a). The same noise was achieved in dual-source CT, even though twice the total tube current–time product was applied, because only a subset of each tube's projections are used in dual-source CT image formation to achieve the decreased temporal resolution (Table 2). Hence, the final dose values were higher for dual-source CT (23.7 mGy for multi–detector row CT and 35.0 mGy for dual-source CT).


Figure 8A
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8a: Noise expressed as the standard deviation of pixel values in Hounsfield units (left axis) versus tube current–time product per tube per rotation and volume CT dose index (CTDIvol) in milligrays (right axis) versus tube current–time product per tube per rotation for multi–detector row CT (MDCT) and dual-source CT (DSCT) for (a) spiral coronary artery calcium protocols and (b) spiral coronary CT angiographic protocols. For dual-source CT, the tube current–time product per tube per rotation corresponds to the tube current–time product value used on each tube. Doses for each examination with multi–detector row CT and dual-source CT are determined at equivalent noise values (horizontal lines at 20 HU for coronary artery calcium protocol and 25 HU for CT angiographic protocol).

 

Figure 8B
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8b: Noise expressed as the standard deviation of pixel values in Hounsfield units (left axis) versus tube current–time product per tube per rotation and volume CT dose index (CTDIvol) in milligrays (right axis) versus tube current–time product per tube per rotation for multi–detector row CT (MDCT) and dual-source CT (DSCT) for (a) spiral coronary artery calcium protocols and (b) spiral coronary CT angiographic protocols. For dual-source CT, the tube current–time product per tube per rotation corresponds to the tube current–time product value used on each tube. Doses for each examination with multi–detector row CT and dual-source CT are determined at equivalent noise values (horizontal lines at 20 HU for coronary artery calcium protocol and 25 HU for CT angiographic protocol).

 
With the dual-source CT system and the B26 kernel used for coronary CT angiographic examinations, a three-dimensional adaptive noise reduction was employed, whereas with the multi–detector row CT system and the B25 kernel used for the same examinations, only a two-dimensional adaptive noise reduction scheme was used. Hence, the same image noise was achieved at a lower tube current–time product setting for each tube of the dual-source CT system (Fig 8b). Hence, the final dose values were relatively comparable (58.9 mGy for multi–detector row CT and 61.2 mGy for dual-source CT). These values of dose for an equivalent noise were for a pitch of 0.2 for both systems and did not include electrocardiographically based tube current modulation.

Dose as a Function of Pitch and Electrocardiographic Modulation
As heart rate increased, the increase in pitch provided the most dramatic dose reductions (Fig 9, Table 5). Tube current modulation was more efficient at dose reduction at lower heart rates compared with that at higher heart rates.


Figure 9
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9: Cumulative dose reduction for coronary CT angiography obtained by using the four dose-reduction mechanisms implemented with the dual-source CT (DSCT) system. The black and gray bars are for the situation in which electocardiographically dependent tube current modulation was not used. Note that dose decreases both with increasing pitch (pitch of 0.2–0.46 for dual-source CT) and with decreasing width of the maximum tube current temporal window (400–110 msec). Electrocardiographically based tube current modulation is somewhat less efficient at higher heart rates, which is demonstrated most clearly by the multi–detector row CT (MDCT) data where pitch must be held constant at 0.2 as heart rate increases. BPM = beats per minute, CTDIvol = volume CT dose index.

 

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

  		Table 5. Volume CT Dose Index for Spiral Multi–Detector Row CT and Dual-Source CT Coronary Artery Angiography

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
IMPLICATION FOR PATIENT CARE
 References
 
The data presented here demonstrate that the simultaneous operation of two x-ray sources need not contribute to an increase in the dose for electrocardiographically gated cardiac CT examinations; rather, the doses from electrocardiographically gated coronary CT angiographic examinations of equivalent noise can be reduced by up to a factor of two relative to single-source multi–detector row CT, depending on the heart rate of the patient. The 88% increase in dose relative to multi–detector row CT when both dual-source CT tubes were simultaneously energized was mitigated through the use of four dose reduction mechanisms: a targeted FOV cardiac beam-shaping filter, three-dimensional adaptive noise reduction, increased pitch values for higher heart rates, and electrocardiographically based tube current modulation with narrow temporal windows and fast tube current transitions. Although the first two of these methods can be effectively implemented with single-source CT systems, the effectiveness of increasing pitch and of tube current modulation are dependent on the improved temporal resolution of the dual-source CT system.

The capability to raise pitch to values as high as 0.5 for heart rates of more than 100 beats per minute is a direct consequence of the 83-msec temporal resolution of the dual-source CT system that requires only single-segment reconstructions at all heart rates. In single-source cardiac CT, motion artifacts would be unacceptable at these heart rates without the use of multisegment reconstruction methods, which require the use of projection data from two or more cardiac cycles to improve the effective temporal resolution. Such techniques require a relatively slow table speed that corresponds to pitch values of approximately 0.2 for scanners with gantry rotation times shorter than 0.4 second. With dual-source CT, multisegment reconstruction methods are not needed, and the pitch value can be increased. This capability to increase pitch translates into a linear decrease in patient dose. Although some scanners do allow changes in pitch as a function of patient heart rate, the increase in pitch is rather small (eg, from 0.2 to 0.24).

Although electrocardiographically based tube current modulation has been shown to reduce dose by 30%–50% for single-source systems (27), our experience is that few sites routinely use it for their cardiac CT angiographic examinations. The reason is that, even with 165-msec temporal resolution, the optimal reconstruction phase (the one with the least motion artifact) can vary by several hundred milliseconds between end systole and end diastole (14,15). If the optimal phase occurs at a point at which the tube current has been decreased, the image quality will not be sufficient. With dual-source CT, the 83-msec temporal resolution demonstrates negligible motion artifact in the coronary arteries at most phases within the cardiac cycle (20,28,29), and this factor should lead to routine use of electrocardiographic tube current modulation.

A limitation of this study is that dual-source CT was compared with only one multi–detector row CT system, a system that was from the same manufacturer. This factor, however, minimizes the potential sources of variation between the systems, because they have the same data acquisition systems, x-ray tubes, and reconstruction algorithms. Comparisons between doses with other CT systems can be readily performed by using the volume CT dose index values provided in this work and those reported on other systems' operator consoles according to International Electrotechnical Commission standards in publication 60601-2-44 (22).

A second limitation of this study is that an assumption was made about the appropriate width of the temporal window in dual-source CT for clinical acceptance. The temporal window for images created by using a partial scan reconstruction can be as narrow as 110 msec because of the improved temporal resolution of the system. Our early clinical use of the system indicates that this window is sufficient for patients with lower heart rates but that a wider window is desirable for patients with higher heart rates in order to allow some optimization of the reconstruction phase. When we allowed for this limitation by using a window of approximately 310 msec, we observed that a dose reduction of a factor of two can be achieved for dual-source CT cardiac CT examinations in patients with higher heart rates; as a result, images with the same noise and spatial resolution as for multi–detector row CT are produced, yet with a factor of two improved temporal resolution. The best compromise between dose reduction and clinical robustness will need to be determined with further clinical studies in order to define the most appropriate temporal window width.

A final limitation of this work is the assumption that the temporal resolution of the system (83 msec) will be adequate for all heart rates and in the presence of heart rate variability. If further improvements in temporal resolution are found to be clinically necessary for higher and varying heart rates, multisector reconstruction approaches might also be needed for dual-source CT. In this case, the expected dose reductions will be decreased as a consequence of the need to maintain lower pitch values. To address the clinical acceptability of 83-msec temporal resolution of dual-source CT in a broad range of patients, clinical evaluations are required. At present, researchers in two preliminary studies conclude that the technology offers robust image quality independent of the heart rate (28,29).


   IMPLICATION FOR PATIENT CARE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 IMPLICATION FOR PATIENT CARE
References
 


   FOOTNOTES
 

Abbreviations: FOV = field of view

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, C.H.M., T.G.F.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, C.H.M., A.N.P., O.S., H.B., K.S., R.R., B.S., B.M.O., T.G.F.; experimental studies, C.H.M., A.N.P., O.S., H.B., K.S., R.R., C.S., B.S., T.G.F.; and manuscript editing, C.H.M., A.N.P., O.S., H.B., B.S., T.G.F.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 IMPLICATION FOR PATIENT CARE
 References
 

  1. Ohnesorge B, Flohr T, Becker C, et al. Cardiac imaging by means of electrocardiographically gated multisection spiral CT: initial experience. Radiology 2000;217:564–571.[Abstract/Free Full Text]
  2. Kachelriess M, Ulzheimer S, Kalender WA. ECG-correlated image reconstruction from subsecond multi-slice spiral CT scans of the heart. Med Phys 2000;27:1881–1902.[CrossRef][Medline]
  3. Taguchi K, Anno H. High temporal resolution for multislice helical computed tomography. Med Phys 2000;27:861–872.[CrossRef][Medline]
  4. Morin RL, Gerber TC, McCollough CH. Radiation dose in computed tomography of the heart. Circulation 2003;107:917–922.[Free Full Text]
  5. McCollough CH. Patient dose in cardiac computed tomography. Herz 2003;28:1–6.[CrossRef][Medline]
  6. 1990 Recommendations of the International Commission on Radiological Protection. Ann ICRP 1991;21:1–201.[Medline]
  7. National Council on Radiation Protection and Measurements. Ionizing radiation exposure of the population of the United States. Bethesda, Md: National Council on Radiation Protection and Measurements, 1987.
  8. Flohr T, Ohnesorge B, Bruder H, et al. Image reconstruction and performance evaluation for ECG-gated spiral scanning with a 16-slice CT system. Med Phys 2003;30:2650–2662.[CrossRef][Medline]
  9. Flohr TG, Schoepf UJ, Kuettner A, et al. Advances in cardiac imaging with 16-section CT systems. Acad Radiol 2003;10:386–401.[CrossRef][Medline]
  10. Flohr TG, Stierstorfer K, Ulzheimer S, Bruder H, Primak AN, McCollough CH. Image reconstruction and image quality evaluation for a 64-slice CT scanner with z-flying focal spot. Med Phys 2005;32:2536–2547.[CrossRef][Medline]
  11. Primak AN, McCollough CH, Bruesewitz MR, Zhang J, Fletcher JG. Relationship between noise, dose, and pitch in cardiac multi-detector row CT. RadioGraphics 2006;26:1785–1794.[Abstract/Free Full Text]
  12. Flohr T, Ohnesorge B. Heart rate adaptive optimization of spatial and temporal resolution for electrocardiogram-gated multislice spiral CT of the heart. J Comput Assist Tomogr 2001;25:907–923.[CrossRef][Medline]
  13. Leber AW, Knez A, von Ziegler F, et al. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol 2005;46:147–154.[Abstract/Free Full Text]
  14. Raff GL, Gallagher MJ, O'Neill WW, Goldstein JA. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 2005;46:552–557.[Abstract/Free Full Text]
  15. Mollet NR, Cademartiri F, van Mieghem CA, et al. High-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography. Circulation 2005;112:2318–2323.[Abstract/Free Full Text]
  16. Ohnesorge BM, Westerman BR, Schoepf UJ. Scan techniques for cardiac and coronary artery imaging with multislice CT. In: Schoepf UJ, ed. CT of the heart: principles and applications. Totowa, NJ: Humana, 2005; 23–43.
  17. Halliburton SS, Stillman AE, Flohr T, et al. Do segmented reconstruction algorithms for cardiac multi-slice computed tomography improve image quality? Herz 2003;28:20–31.
  18. Robb RA, Ritman EL. High-speed synchronous volume computer tomography of the heart. Radiology 1979;133:655–661.[Abstract]
  19. Ritman EL, Kinsey JH, Robb RA, Gilbert BK, Harris LD, Wood EH. Three-dimensional imaging of heart, lungs, and circulation. Science 1980;210:273–280.[Abstract/Free Full Text]
  20. Flohr TG, McCollough CH, Bruder H, et al. First performance evaluation of a dual-source CT (DSCT) system. Eur Radiol 2006;16:256–268.[CrossRef][Medline]
  21. Schardt P, Deuringer J, Freudenberger J, et al. New x-ray tube performance in computed tomography by introducing the rotating envelope tube technology. Med Phys 2004;31:2699–2706.[CrossRef][Medline]
  22. International Electrotechnical Commission. Medical electrical equipment: part 2-44—particular requirements for the safety of x-ray equipment for computed tomography. Publication no. 60601-2-44. Ed. 2.1. Geneva, Switzerland: International Electrotechnical Commission, 2002; 1–36.
  23. McNitt-Gray MF. AAPM/RSNA physics tutorial for residents: topics in CT—radiation dose in CT. RadioGraphics 2002;22:1541–1553.[Abstract/Free Full Text]
  24. European Commission. European guidelines for quality criteria for computed tomography. Luxembourg: European Commission, 2000.
  25. Raupach R, Bruder H, Stierstorfer K, Suess C, Flohr T. A novel approach for efficient edge preserving noise reduction in CT volume data [abstr]. In: Radiological Society of North America Scientific Assembly and Annual Meeting Program. Oak Brook, Ill: Radiological Society of North America, 2005; 524.
  26. Ulzheimer S, Kalender WA. Assessment of calcium scoring performance in cardiac computed tomography. Eur Radiol 2003;13:484–497.[Medline]
  27. Jakobs TF, Becker CR, Ohnesorge B, et al. Multislice helical CT of the heart with retrospective ECG gating: reduction of radiation exposure by ECG-controlled tube current modulation. Eur Radiol 2002;12:1081–1086.[CrossRef][Medline]
  28. Johnson TR, Nikolaou K, Wintersperger BJ, et al. Dual-source CT cardiac imaging: initial experience. Eur Radiol 2006;16:1409–1415.[CrossRef][Medline]
  29. Achenbach S, Ropers D, Kuettner A, et al. Contrast-enhanced coronary artery visualization by dual-source computed tomography: initial experience. Eur J Radiol 2006;57:331–335.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
P. Stolzmann, S. Leschka, H. Scheffel, T. Krauss, L. Desbiolles, A. Plass, M. Genoni, T. G. Flohr, S. Wildermuth, B. Marincek, et al.
Dual-Source CT in Step-and-Shoot Mode: Noninvasive Coronary Angiography with Low Radiation Dose
Radiology, October 1, 2008; 249(1): 71 - 80.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
P. Stolzmann, T. Frauenfelder, T. Pfammatter, N. Peter, H. Scheffel, M. Lachat, B. Schmidt, B. Marincek, H. Alkadhi, and T. Schertler
Endoleaks after Endovascular Abdominal Aortic Aneurysm Repair: Detection with Dual-Energy Dual-Source CT
Radiology, September 9, 2008; (2008) 2483080193.
[Abstract] [Full Text]

Home page
 RadiologyHome page
A. C. Weustink, N. R. Mollet, F. Pugliese, W. B. Meijboom, K. Nieman, M. H. Heijenbrok-Kal, T. G. Flohr, L. A. E. Neefjes, F. Cademartiri, P. J. de Feyter, et al.
Optimal Electrocardiographic Pulsing Windows and Heart Rate: Effect on Image Quality and Radiation Exposure at Dual-Source Coronary CT Angiography
Radiology, September 1, 2008; 248(3): 792 - 798.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
B. K. Courtney, N. R. Munce, K. J. Anderson, A. S. Thind, G. Leung, P. E. Radau, F. S. Foster, I. A. Vitkin, R. S. Schwartz, A. J. Dick, et al.
Innovations in imaging for chronic total occlusions: a glimpse into the future of angiography's blind-spot
Eur. Heart J., March 1, 2008; 29(5): 583 - 593.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J. P. Earls, E. L. Berman, B. A. Urban, C. A. Curry, J. L. Lane, R. S. Jennings, C. C. McCulloch, J. Hsieh, and J. H. Londt
Prospectively Gated Transverse Coronary CT Angiography versus Retrospectively Gated Helical Technique: Improved Image Quality and Reduced Radiation Dose
Radiology, March 1, 2008; 246(3): 742 - 753.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
P. Schoenhagen
Back to the future: coronary CT angiography using prospective ECG triggering
Eur. Heart J., January 2, 2008; 29(2): 153 - 154.
[Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
H. Seifarth, S. Wienbeck, M. Pusken, K.-U. Juergens, D. Maintz, C. Vahlhaus, W. Heindel, and R. Fischbach
Optimal Systolic and Diastolic Reconstruction Windows for Coronary CT Angiography Using Dual-Source CT
Am. J. Roentgenol., December 1, 2007; 189(6): 1317 - 1323.
[Abstract] [Full Text] [PDF]

Home page
 RadioGraphicsHome page
D. D. Cody and M. Mahesh
AAPM/RSNA Physics Tutorials AAPM/RSNA Physics Tutorial for Residents: Technologic Advances in Multidetector CT with a Focus on Cardiac Imaging
RadioGraphics, November 1, 2007; 27(6): 1829 - 1837.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
U. J. Schoepf, P. Suranyi, C. Thilo, P. Costello, and P. L. Zwerner
Replacing diagnostic catheterization with coronary CT angiography: the final frontier
Eur. Heart J., October 1, 2007; 28(19): 2305 - 2306.
[Full Text] [PDF]

Home page
 CirculationHome page
A. J. Einstein, K. W. Moser, R. C. Thompson, M. D. Cerqueira, and M. J. Henzlova
Radiation Dose to Patients From Cardiac Diagnostic Imaging
Circulation, September 11, 2007; 116(11): 1290 - 1305.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2433061165v1
243/3/775    most recent
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