(Radiographics. 2000;20:1675-1681.)
© RSNA, 2000
Motion Artifacts in Subsecond Conventional CT and Electron-Beam CT: Pictorial Demonstration of Temporal Resolution1
Cynthia H. McCollough, PhD,
Michael R. Bruesewitz, RT(R),
Timothy R. Daly, RT(R) and
Frank E. Zink, PhD
1 From the Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. Recipient of a Certificate of Merit award for a scientific exhibit at the 1999 RSNA scientific assembly. Received February 25, 2000; revision requested April 24 and received June 19; accepted June 19. Address correspondence to C.H.M. (e-mail: mccollough.cynthia@mayo.edu).
 |
Abstract
|
|---|
To visually demonstrate the effective temporal resolution of subsecond conventional (slip-ring) and electron-beam computed tomographic (CT) systems, two phantoms containing high-contrast test objects were scanned with a slip-ring CT system (effective exposure time, 0.5 second) and an electron-beam CT system (exposure time, 0.1 second). Images were acquired of each phantom at rest, during translation along the x axis at speeds of 10–100 mm/sec, and during rotation about isocenter at speeds of 0.1 and 0.5 revolution per second. Motion artifacts and loss of spatial resolution were judged to be absent, noticeable, or severe. For 0.5-second conventional CT images, motion artifacts and loss of spatial resolution were noticeable at 10 mm/sec and 0.1 revolution per second and were severe at speeds greater than or equal to 20 mm/sec and at 0.5 revolution per second. For 0.1-second electron-beam CT scans, noticeable, but not severe, motion artifacts and loss of spatial resolution occurred at speeds between 40 and 100 mm/sec and at 0.5 revolution per second. Over the range of physiologic speeds examined, the images provide visually compelling evidence of the effect of improving temporal resolution in CT.
Index Terms: Computed tomography (CT), artifact • Computed tomography (CT), electron beam • Computed tomography (CT), image quality • Phantoms • Test objects
|
Introduction
|
|---|
It is well established that patient motion during scan acquisition causes artifacts in computed tomographic (CT) images and that physiologic motion is particularly problematic. Since the introduction of CT in 1973, acquisition time per single image acquisition has continued to fall. Currently, conventional CT scanners can acquire a single image in times as short as 0.8, 0.75, or 0.5 second for 360° of projection data. Electron-beam CT scanners can acquire a single image in times as short as 0.1 or 0.05 second for 240° of projection data. Still, these short exposure times may not be sufficient for artifact-free imaging of cardiac structures and the surrounding tissue (1,2).
The Table summarizes measured velocities for several cardiac and pulmonary structures (3,4). Ritchie et al (3) concluded that object displacements of approximately 1 pixel width can introduce significant image artifact. However, such estimates are dependent on several factors, including (but not limited to) the number of projection angles used in the image reconstruction; the use of motion-reduction reconstruction algorithms; the definition of "acceptable" artifact; the direction of motion; the size and attenuation of the object; and the geometry of the scanner. Rather than relying on mathematical simulations, we sought to empirically determine the amount of artifact created by translational and rotational motion for two known, state-of-the-art, short scan acquisition times. An important goal of our work was to demonstrate the appearance and extent of motion artifacts created by test phantoms having physiologic velocities within the ranges noted in the Table. In this manner, users of "fast" CT systems can gain a better appreciation of the degree of expected motion artifact as a function of object velocity.
|
Materials and Methods
|
|---|
The motion phantom used in this study (Fig 1) was designed and constructed at our institution. The drive mechanism can be used to translate or rotate a phantom that has been mounted on the apparatus. Two test objects were examined: a high-contrast spatial resolution phantom (Fig 2) and an acrylic cylinder containing three regions of tissue-equivalent bolus material with embedded high-contrast test objects (Fig 3). Two of the cylinders of the bolus material contained only air bubbles. The central cylinder contained 2–3-mm pieces of eggshell embedded in a household glue stick.
CT images were obtained with a conventional (slip-ring) CT scanner (CT/i; GE Medical Systems, Milwaukee, Wis) (120 kVp, 60 mA, 0.8-second exposure with partial reconstruction [effective exposure time, 0.5 second], 3-mm scan width, standard reconstruction algorithm) and an electron-beam CT scanner (C150L-XP; Imatron, South San Francisco, Calif) (130 kVp, 620 mA, 0.1-second exposure, 3-mm scan width, sharp reconstruction algorithm). Scanner technique factors were prospectively chosen to give comparable image noise. To confirm that image noise was comparable, noise values were measured within uniform regions of interest in both acrylic and water portions of the phantoms. To isolate in-plane temporal resolution, table motion during scan acquisition (ie, helical or continuous volume scanning) was not used.
Images of each phantom were acquired at rest; during translation along the x axis (left-right) at speeds of 10, 20, 30, 40, 50, 60, and 100 mm/sec; and during rotation about isocenter at speeds of 0.1 and 0.5 revolution per second. Motion artifacts and loss of spatial resolution were judged to be absent, noticeable, or severe. Artifacts not related to motion (eg, partial volume artifacts of the air bubbles that made them appear noncircular) were ignored in the grading of the images. Because a subjective classification scheme was used, what one reader might score as noticeable another might consider absent. However, despite subtle differences in classification assignment, the scheme was useful in assessing overall trends. The reader is encouraged to determine for himself or herself the degree to which the various motion artifacts degraded image fidelity.
When viewed from the foot of the patient table (looking toward the front of the gantry), the x-ray beam moved in a clockwise direction for both the conventional scanner and the electron-beam scanner. The patient orientation (head first vs feet first) was different between the two scanners for data acquired from the resolution target. Thus, the left-right orientation of the resolution target was reversed between the two scanners. As a consequence, the direction of phantom rotation (when viewed from the foot of the patient table) was clockwise for the conventional scanner and counterclockwise for the electron-beam scanner. For data acquired with the air bubble and eggshell phantom, the patient orientations were the same and the objects rotated in the counterclockwise direction.
|
Results
|
|---|
The technique factors chosen resulted in comparable image noise (15–19 HU for conventional CT, 23–26 HU for electron-beam CT). Noticeable motion artifacts and loss of spatial resolution occurred at 10 mm/sec and 0.1 revolution per second for 0.5-second conventional CT scans. These effects became severe at speeds greater than or equal to 20 mm/sec and at 0.5 revolution per second (Figs 4, 5). Noticeable, but not severe, motion artifacts and loss of spatial resolution occurred at speeds between 40 and 100 mm/sec and at 0.5 revolution per second for 0.1-second electron-beam CT scans (Figs 6, 7).
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4d. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4e. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4f. Conventional CT images (0.5-second effective exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows noticeable artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows severe artifacts.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (164K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5e. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (162K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5f. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5g. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5h. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows noticeable artifacts. (c, d) Images obtained during 20-mm/sec (c) and 40-mm/sec (d) linear motion show severe artifacts. (e, f) Images obtained during 0.1-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter). (g, h) Images obtained during 0.5-revolution-per-second rotational motion show severe artifacts.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6d. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6e. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6f. Electron-beam CT images (0.1-second exposure time) of the resolution phantom. (a) Image obtained at rest shows an absence of artifacts. (b) Image obtained during 10-mm/sec linear motion shows an absence of artifacts. (c, d) Images obtained during 40-mm/sec (c) and 100-mm/sec (d) linear motion show noticeable artifacts. (e) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (f) Image obtained during 0.5-revolution-per-second rotational motion shows noticeable artifacts.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7d. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7e. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7f. Electron-beam CT images (0.1-second exposure time) of the air bubble and eggshell phantom. (a) Image obtained at rest shows an absence of artifacts. (b, c) Images obtained during 40-mm/sec (b) and 100-mm/sec (c) linear motion show noticeable artifacts. (d) Image obtained during 0.1-revolution-per-second rotational motion shows an absence of artifacts. (e, f) Images obtained during 0.5-revolution-per-second rotational motion show noticeable or severe artifacts (depending on the distance from isocenter).
|
|
The degree and appearance of artifacts for the conventional scanner were dependent on the direction of the translational movement (Fig 8). A similar example is noted for rotational movement in Figure 5e–5h. We hypothesize that this dependence is due to the random correlation of the position of the x-ray tube (and hence primary projection data) with the phantom motion.
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom obtained during 50-mm/sec linear motion. Images obtained while the phantom was moving from left to right (a) and from right to left (b) show the dependence of artifact appearance on the direction of translational movement.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Conventional CT images (0.5-second effective exposure time) of the air bubble and eggshell phantom obtained during 50-mm/sec linear motion. Images obtained while the phantom was moving from left to right (a) and from right to left (b) show the dependence of artifact appearance on the direction of translational movement.
|
|
|
Discussion
|
|---|
Neither left-right nor rotational movement (clockwise or counterclockwise) adequately represents any specific physiologic motion. However, the design and construction of a phantom that is a more realistic model of patient motion is a complex and difficult undertaking. Although such a tool may further demonstrate the specific appearance of artifacts created by a specific motion, the translational and rotational motions of the phantom used are adequate for demonstration of the severity of motion artifacts as a function of scan time. Translational motion produced gross artifacts on the exterior of the phantom due to the remarkable shift of the object periphery during scan acquisition. This result might correlate with large shifts in peripheral patient structures due to breathing or other large motions. The rotational phantom, which is circularly symmetric at the periphery, did not create these gross surface artifacts and better represents the situation of a beating heart or other internal motion that does not include a shift in the patient periphery. Thus, the two motions provide complementary information.
Furthermore, rotational motion is a reasonable model of the in-plane motion of the coronary arteries, which move along an arc tangential to both the x and y axes. Since the coronary arteries move along the arc and return (rather than rotate continuously about some center point), the rotational velocities were chosen so that the phantoms rotated by less than one complete revolution per scan. On the conventional scanner (effective scan time of 0.5 second), the phantom rotated approximately one-20th and one-fourth of a revolution per scan acquisition. On the electron-beam scanner (effective scan time of 0.1 second), the phantom rotated approximately one-100th and one-20th of a revolution per scan acquisition. The images visually demonstrate the principle that for objects moving at the same velocity, images acquired with shorter exposure times record a smaller net movement of the object and hence are subject to fewer reconstruction artifacts.
An absolute velocity (in millimeters per second) of the air bubbles and eggshells during rotational motion can be estimated with knowledge of the rotational velocity (eg, 0.5 revolution per second) and the circumference at any given radius (2
R, where R is the radius of the object of interest). For the eggshell fragments, which were at a very small radius from isocenter (
5 mm), the velocity at 0.5 revolution per second was 16 mm/sec, which is considerably slower than most of the maximum velocities noted in the Table. For the air bubbles, which were at a larger radius from isocenter (30–50 mm), the velocities at 0.5 revolution per second were 94–157 mm/sec, which are similar to the maximum velocities of the coronary arteries (Table). This dependence of rotational object velocity on the radius of the object of interest is visually demonstrated by the increasing severity of motion artifacts from the center of rotation toward the object periphery.
|
Conclusions
|
|---|
As expected, shorter acquisition times produce fewer artifacts for the same object velocity due primarily to the decreased object displacement. For scans with the same object displacement (10-mm displacement occurs at 20 mm/sec for a 0.5-second conventional CT exposure or at 100 mm/sec for a 0.1-second electron-beam CT exposure), differences in the degree and appearance of motion artifacts are small (Fig 4c vs Fig 6d, Fig 4e vs Fig 6f, Fig 5c vs Fig 7c, Fig 5e and 5f vs Fig 7e and 7f). Hence, the appearance of motion artifacts at scan times not evaluated in our study can be estimated from images with object displacements similar to those we have studied.
Over the range of physiologic speeds examined, the images provide visually compelling evidence of the effect of improving temporal resolution in CT. Similar conclusions have been drawn from in vivo experiments (5,6). Recently introduced conventional scanners with 0.5-second gantry rotation times (0.3-second effective scan time for partial reconstruction) will extend the abilities of conventional CT at higher speeds. However, the data presented herein indicate that, for linear velocities between 20 and 100 mm/sec and rotational velocities of 0.5 revolution per second, scan times of 0.1 second are required to avoid severe motion artifacts.
|
Footnotes
|
|---|
Nothing in this publication implies that Mayo Foundation endorses the products used in this study.
|
References
|
|---|
-
Mao S, Lu B, Oudiz RJ, et al. Coronary artery motion in electron beam tomography. J Comput Assist Tomogr 2000; 24:253-258.[Medline]
-
Achenbach S, Ropers D, Holle J, et al. In-plane coronary arterial motion velocity: measurement with electron-beam CT. Radiology 2000; 216:457-463.[Abstract/Free Full Text]
-
Ritchie CJ, Godwin JD, Crawford CR, et al. Minimum scan speeds for suppression of motion artifacts in CT. Radiology 1992; 185:37-42.[Abstract/Free Full Text]
-
Hofman MB, Wickline SA, Lorenz CH. Quantification of in-plane motion of the coronary arteries during the cardiac cycle: implications for acquisition window duration for MR flow quantification. J Magn Reson Imaging 1998; 8:568-576.[Medline]
-
Schoepf UJ, Helmberger T, Holzknecht N, et al. Segmental and subsegmental pulmonary arteries: evaluation with electron-beam versus spiral CT. Radiology 2000; 214:433-439.[Abstract/Free Full Text]
-
Reinhardt JM, Hoffman EA. Quantitative pulmonary imaging: spatial and temporal considerations in high-resolution CT. Acad Radiol 1998; 5:539-546.[Medline]
This article has been cited by other articles:
|
|
|
|
 
R. Popilock, K. Sandrasagaren, L. Harris, and K. A. Kaser
CT Artifact Recognition for the Nuclear Technologist
J. Nucl. Med. Technol.,
June 1, 2008;
36(2):
79 - 81.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Grosjean, B. Sauer, R. M. Guerra, M. Daudon, A. Blum, J. Felblinger, and J. Hubert
Characterization of Human Renal Stones with MDCT: Advantage of Dual Energy and Limitations Due to Respiratory Motion
Am. J. Roentgenol.,
March 1, 2008;
190(3):
720 - 728.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. Liu, C. Cuevas, A. A. Moss, O. Kolokythas, T. J. Dubinsky, and P. E. Kinahan
Gas Bubble Motion Artifact in MDCT
Am. J. Roentgenol.,
February 1, 2008;
190(2):
294 - 299.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. S. Choi, B. W. Choi, K. O. Choe, D. Choi, K.-J. Yoo, M.-I. Kim, and J. Kim
Pitfalls, Artifacts, and Remedies in Multi- Detector Row CT Coronary Angiography
RadioGraphics,
May 1, 2004;
24(3):
787 - 800.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
