| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Radiology, Emory University School of Medicine, 1364 Clifton Rd NE, Atlanta, GA 30322. Received January 14, 2006; revision requested March 22; revision received March 27; accepted May 2; final version accepted May 19. Address correspondence to S.S. (e-mail: ssaini{at}emory.edu).
Numerous articles have been published on the evaluation of computed tomographic (CT) technology and its applications. However, to our knowledge, there are no specific guidelines or recommendations for describing CT technique, which is important because of the magnitude of clinical applications and the contribution of CT to radiation dose (1). In this editorial, we will discuss the need for a standardized nomenclature and a uniform presentation of CT scanning parameters and contrast material administration protocols. We will also discuss the reasons for the lack of such standardization in scientific articles and propose a systematic method of standardization.
 |
NEED FOR STANDARDIZATION |
|---|
 TOPINTRODUCTION NEED FOR STANDARDIZATION REASONS FOR LACK FOR...PROPOSALS FOR STANDARDIZATIONSUMMARYReferences |
|---|
First, the use of conflicting nomenclature and inconsistent or incomplete descriptions of scanning parameters can affect the reproducibility of validation studies performed in different settings or centers. Validation studies are important components of the scientific assessment of any given technique or clinical trial.
Second, the lack of a standardized presentation of the scanning protocols can convey inappropriate information to radiologists, technologists, or medical physicists, who are responsible for developing the CT protocols for their scanners. The use of a standardized nomenclature and uniform description of scanning protocols may help create homogeneity between scanning protocols that are used at different CT centers. This is an important aspect of multi–detector row CT scanning because of the large variability among different CT centers (2).
Because a standardized nomenclature can help researchers avoid the use of several confusing proprietary terms, it may become easier for radiologists to adapt the various scanning protocols presented in scientific articles to their clinical practice. Furthermore, with the constant evolution of multi–detector row CT technology, newer applications are being described and involve a greater degree of complexity that may be simplified with standardization (3).
Finally, we believe that a standardized nomenclature and uniform presentation of CT scanning protocols can help radiology personnel estimate the radiation dose that is associated with CT.
Unfortunately, despite an increasing consensus over the risks associated with radiation dose, the authors of many scientific articles on CT (with the exception of articles related to radiation dose) do not explicitly mention the radiation dose that is associated with their protocols. In this respect, standardization can help readers to compare published scanning protocols with their own protocols and, if needed, to adapt their protocols according to those in a peer-reviewed study for managing radiation dose.
It is important, however, to understand that standardization systems may rapidly become obsolete as a result of the rapidly evolving nature of CT technology.
|
REASONS FOR LACK FOR STANDARDIZATION |
|---|
|
TOPINTRODUCTIONNEED FOR STANDARDIZATIONREASONS FOR LACK FOR...PROPOSALS FOR STANDARDIZATIONSUMMARYReferences |
|---|
Moreover, the use of proprietary terms for similar techniques developed by different vendors also causes confusion, particularly for radiologists who lack experience with or knowledge of the equipment of all vendors. One example is the use of the terms milliampere, milliampere-second, and effective milliampere-second (defined as milliamperes divided by beam pitch) to describe CT tube current or tube current–time product. Vendor-specific descriptions of reconstruction kernels can also cause similar confusion.
Although there are substantial difficulties involved with generating a standardized nomenclature and format for describing scanning techniques and contrast material administration protocols, we believe that this task is achievable with the use of a simple and uniform presentation of scanning parameters and nomenclature.
|
PROPOSALS FOR STANDARDIZATION |
|---|
|
TOPINTRODUCTIONNEED FOR STANDARDIZATIONREASONS FOR LACK FOR...PROPOSALS FOR STANDARDIZATIONSUMMARYReferences |
|---|
Standardized Nomenclature and Uniform Presentation of Scanning Techniques
Although there will always be some disagreement among experienced individuals and vendors on any proposal for uniform nomenclature, we believe that constructive criticism can help the imaging community gain an understanding of complex scanning parameters. We propose the following nomenclature for scanning parameters of routine CT applications: First, we propose that the term multi–detector row CT replace terms such as multislice, multichannel, or multisection CT. In addition, authors should specify whether the imaging technique is single source or dual source, as well as the trade name and manufacturer of the scanner, the number of total data channels (expressed as sections per rotation to include sampling by means of a focal spot that oscillates in the longitudinal direction during x-ray tube rotation [z-flying focal spot]), the software version that was used at the time of study, the patient's orientation with respect to the scanner gantry (head first or feet first), the patient's position on the gantry table (supine, prone, right decubitus, or left decubitus), the number of localizer images acquired (for scout view, topogram, surview, or scanogram), the anatomic area of interest, the scanning direction (cephalocaudad or caudocephalad), breathing command (no breath-hold instruction given, breath-hold instruction given [if given without any specific instruction for inspiratory or expiratory breath hold, such as for acquisition of localizer images], inspiratory breath-hold instruction given, or expiratory breath-hold instruction given), tube potential (in peak kilovoltage), and tube current.
For tube current, it should be specified whether fixed tube current or automatic exposure control was used. Fixed tube current should be expressed in milliamperes and not in milliampere-seconds per section or effective milliampere-seconds. Automatic exposure control should be expressed as angular (for xy- or angle-modulated automatic exposure control techniques), z-axis (for z-modulation, localizer-based, or longitudinal automatic exposure control techniques), or combined xyz (for xyz-modulation or combined-modulation automatic exposure control techniques).
Other details that are pertinent to the study should also be given (4), including noise index and milliampere limits, effective milliampere-second, and reference values for effective milliampere-second and modulation settings in obese and slim patients. Also, the term gantry revolution time (in seconds) should be used instead of misnomer gantry rotation time, and the term table speed (in millimeters per revolution instead of in millimeters per second) should be used instead of table travel. Beam collimation should be expressed as the number of data channels times the effective detector-row width (in millimeters) (eg, 64 x 0.6 mm). Likewise, the term beam pitch, which is defined as the table speed (in millimeters) per gantry revolution divided by beam collimation (in millimeters), should be used instead of pitch, slice pitch, or pitch factor. Likewise, the term reconstructed section thickness (in millimeters) should be used instead of slice thickness, slice width, slice profile, effective slice thickness, slices, or sections. Also provided should be the number of reconstructions obtained per CT examination, as well as the reconstructed section thickness, section overlap, and kernels used for each reconstruction. The term section overlap (in millimeters) should be used instead of slice interval, interslice interval, slice gap, section overlap, or interval, and reconstruction kernel should be used instead of reconstruction algorithm.
For each reconstruction kernel, it should be specified whether a soft-tissue reconstruction kernel (for low-spatial-resolution reconstruction kernels) or lung or bone reconstruction kernel (for high-spatial-resolution reconstruction kernels) was used. The manufacturer of the reconstruction kernel, the number of passes (ie, the specific number of passes; the phase of each pass [unenhanced, arterial, venous, early, or delayed]; the anatomic area covered with each pass; the field of view, technique, or delay [from the start of contrast material injection] used to trigger each pass; and scanning parameters [if they are different]), and the duration of scanning (in seconds) for each pass should be provided.
For cardiac CT applications and dynamic motion, details of electrocardiographic gating and/or pulsing, reconstruction intervals, and the use of a ß blocker (number, amount, time, route of administration, and dose [in milligrams]) should be given.
For kinetic CT of the extremities, authors should give the motion extent, axis (in degrees), and scanning coverage in addition to the aforementioned details (the scan description must include protocol-specific techniques). Also, information on radiation dose (CT dose index volume [in milligrays] and dose length product [in milligray-centimeters]) may be summarized with the details of the scanning parameters. Radiation dose from the acquisition of localizer images must also be included.
Considering the large number of scanning and contrast material parameters, we believe that these CT techniques can be presented in a form that the average reader may find easier to understand for comparison purposes and for adaptation to their scanners if they so desire (Fig 1). Each tabular representation can include the aforementioned uniform protocols and begin with the diagnostic indication for the study.
|
|
For a fixed scan delay, authors should specify the fixed delay (in seconds) from the initiation of contrast material injection. For automatic bolus tracking, authors should specify the site of the region of interest, the start time for the acquisition of the monitoring scan after injection, the interval between each scan, the threshold attenuation for the trigger, and the default trigger time (in case of early threshold enhancement or failure to attain threshold enhancement). For the test bolus technique, authors should specify the volume of contrast material used, the rate of bolus injection, the site of the region of interest, the start time for the acquisition of the monitoring scan after injection, the interval between each scan, the estimated delay and volume of contrast material, and any delay time that was added to the estimated delay.
|
SUMMARY |
|---|
|
TOPINTRODUCTIONNEED FOR STANDARDIZATIONREASONS FOR LACK FOR...PROPOSALS FOR STANDARDIZATIONSUMMARYReferences |
|---|
|
FOOTNOTES |
|---|
See also the editorial by McCollough in this issue.
|
References |
|---|
|
TOPINTRODUCTIONNEED FOR STANDARDIZATIONREASONS FOR LACK FOR...PROPOSALS FOR STANDARDIZATIONSUMMARYReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  K. J. Strauss, M. J. Goske, D. P. Frush, P. F. Butler, and G. Morrison Image Gently Vendor Summit: Working Together for Better Estimates of Pediatric Radiation Dose from CT Am. J. Roentgenol., May 1, 2009; 192(5): 1169 - 1175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. McCollough Standardization in CT Terminology: A Physicist's Perspective Radiology, December 1, 2006; 241(3): 661 - 662. [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
