| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 From the Department of Radiology, MC 2026, University of Chicago, 5841 S Maryland Ave, Chicago, IL 60645 (A.H.D.); and the Department of Radiology, Massachusetts General Hospital, Boston (M.E.Z.). Received July 17, 2003; revision requested September 4; revision received October 27; accepted October 28. Address correspondence to A.H.D. (e-mail: ahdachma@uchicago.edu).
Index terms: Colon, CT, 75.1211 • Colon neoplasms, CT, 75.1211, 75.30 • Colon neoplasms, diagnosis, 75.30 • Colonoscopy, 75.30 • Editorials
The rapid development of computed tomographic (CT) colonography has been accompanied by a marked increase in the number of studies with published findings evaluating the technology, diagnostic performance, and clinical acceptance of this technique. Many of the questions now being posed for CT colonography require evaluation in large multicenter trials, few of which have had their results published (1–3) and some of which are in progress. Published data from these larger trials may not become available for years. In the meantime, many important questions are answerable in the synthesis of the findings from single-institution studies. However, to date, meta-analysis of these smaller investigations has been limited by the variability in the acquisition, reporting, and analysis of data. The purpose of this editorial is to propose a standardization of reporting of prospective CT colonographic examination data for researchers and investigators, to facilitate more meaningful comparison and summary of such data in the literature of the future. At the same time, we hope that this editorial will benefit radiologists who perform CT colonography and wish to measure their own success; some of the factors to consider when analyzing their results are indicated.
A typical meta-analysis, as often performed for clinical drug trials, requires a reasonable uniformity in methods. The difficulties in conducting a statistically valid meta-analysis of data from virtual colonoscopy have been summarized previously (4). In one meta-analysis (5), investigators found only 14 articles about CT colonography that met the criteria for inclusion, yet even within these articles, combining the data is statistically problematic. Although the precise methods of performing CT colonography are legislated by local institutional preference and research aims, we propose minimal requirements for reporting of data. The technical parameters for performing CT colonography will vary with the parameters being investigated, but we also propose general minimal guidelines for performance of the examination. We believe all elements relating to these parameters should also be detailed when the results are published.
PERFORMANCE OF EXAMINATION
Unless a specific parameter is being investigated, a clean well-distended colon with little residual fluid is desirable. A saline cathartic (mixture of monobasic and dibasic sodium phosphate [Phosphosoda; C. B. Fleet, Lynchburg, Va], or magnesium citrate) is preferred to colonic lavage (polyethylene glycol), unless the preference of the gastroenterologist for a same-day optical colonoscopy mandates colonic lavage. Respiratory motion should be minimized by using scan times of less than 30 seconds or by using overlapping breath holds; however, the latter method is less desirable for a seamless three-dimensional endoluminal fly-through. In addition, patients may be offered nasal oxygen therapy when necessary and may be given specific instructions to minimize motion if the patient must breathe (eg, "breathe out as slowly as possible to minimize movement of the abdomen"). Both carbon dioxide and room air, administered mechanically, manually, or by the patient, have been used to achieve good distention. Multi–detector row helical scanning is strongly preferred to single-section helical scanning, with the use of (a) collimation of 3 mm or less and (b) overlapping sections at an interval that is two-thirds or less of the collimation. Pitch or table speed should be adjusted to achieve a scan time of 30 seconds or less.
REPORTING OF DATA
We believe that the following essential information must be contained in CT colonography reports: the prevalence of polyps and masses in the cohort; the by-patient and by-polyp sensitivity, specificity, and positive and negative predictive values, each stratified by lesion size; the location of the lesion; histologic findings; and a detailed description of the method for performing the examination, interpreting the findings from the examination, and matching the CT and colonoscopic results. Each of these key items will be justified and explained in the following paragraphs.
Definition of the Cohort Studied
Investigators should report the percentages of asymptomatic average-risk and symptomatic above-average-risk patients in their study cohorts. Ideally, investigators should list the specific symptoms and risk factors (anemia, melena, family history, or known reported risk factors [6]). Newly recognized risk factors, such as association with ovarian and endometrial carcinoma in women (7), should be sought.
The results of clinical trials of other modalities, such as mammography, have shown that diagnostic performance may vary with disease prevalence (8); thus, the cohort must also be defined by the number of polyps and the number of patients with polyps. Because the goal of colorectal cancer screening with CT colonography is the proper triage of patients to optical colonoscopy for confirmation and biopsy of suspected neoplastic lesions, the most clinically salient unit of analysis is the by-patient analysis. Nevertheless, data from a by-polyp analysis should also be provided. A by-polyp analysis is important to understand the capability of CT colonography and because optical colonoscopy is not a perfect reference standard; with optical colonoscopy, as many as 25% of diminutive polyps (<5 mm in diameter) and 6% of polyps 1 cm or larger (9) may be missed. If there is a discrepancy between the findings from CT colonography and optical colonoscopy in a clinical setting, this discrepancy should be evaluated on a case-by-case basis.
In some research settings, discrepancies between CT colonographic and optical colonoscopic findings are resolved either with performance of a second colonoscopy or with segmental unblinding of the colonoscopist during removal of the endoscope. The latter requires that CT colonographic findings are revealed incrementally to the endoscopist during optical colonoscopy as the colonoscope is withdrawn segment by segment, after the endoscopist has recorded his or her own independent observations. In some trials, the endoscopist must be one or even two (10) segments distal to the polyp reported at CT colonography before being unblinded. We strongly advocate that investigators adopt this method to resolve discrepancies between the findings of the two examinations because it is safe and is the potentially most accurate reference standard. However, we recognize that segmental unblinding represents a logistic challenge. The radiologist must provide a same-morning interpretation if a segmental unblinding approach is to be used.
Size of the Lesion
The risk that a given polyp harbors carcinoma appears to be correlated most directly to the size of the lesion. Therefore, both radiologists and endoscopists should make every effort to measure accurately the size of each lesion detected. During endoscopy, the use of a caliper tool is ideal, but this tool is rarely used. Measurement of polyps by comparing them to an open forceps requires that the forceps be held as close as possible to the lesion, a point that should be emphasized as a quality control measure. Thus, polyp size reported at optical colonoscopy is usually at best an estimate. Whatever method is used by the endoscopist should be specified, although in vivo measurements are preferred for comparison with the CT colonographic images. When a polyp is removed in toto, its size may vary from the in vivo size because of an absence of blood flow and the cautery effect. If measured ex vivo, the measurement should be performed before the polyp is placed for a prolonged period in formalin because formalin will shrink the specimen further.
In most cases, the size of the lesion can be measured accurately on magnified two-dimensional multiplanar images, and we recommend that these images should be used in all cases. We recognize that some lesions are oval and perhaps measured more easily on three-dimensional endoluminal views, but three-dimensional measurement of size is subject to potential distortion, and this topic deserves further research. We propose that for standardization, the largest single dimension should be reported for a given lesion, analogous to the "response to treatment in solid tumors criteria" (11). At minimum, investigators should specify how they decided to obtain the measurements and should set rules for uniformity within their own study and between the radiologists and the endoscopists.
To further promote standardization of reporting, we suggest that investigators should report the size of polyps on the basis of standard size categories. The size distributions of reported polyps have been a particularly frustrating issue in attempting to summarize reports because there has been inconsistency in the size groupings used for comparison. Because large masses have a different conspicuity and clinical importance compared with polyps, we suggest that annular cancers and masses 4.0 cm or larger in greatest dimension should be reported and analyzed separately from polyps. Submucosal lipomas should be excluded.
In addition, in some published results, it is unclear how diminutive polyps (<5 mm) have been categorized. Because the incidence of malignancy is extremely low for lesions smaller than 5 mm (except for patients with polyposis syndromes or inflammatory bowel disease, who are usually excluded from the CT colonographic cohorts) and because all investigators have shown poor sensitivity and specificity for lesions smaller than 5 mm, we advise against taking the time to measure and analyze lesions smaller than 5 mm. There is considerable controversy with regard to the 5.0–9.9-mm range, in particular with regard to what size threshold represents an optimal balance between performance and clinical relevance. We therefore suggest that investigators show how the sensitivity, specificity, and positive and negative predictive values would change if a 6-, 7-, 8-, or 9-mm cutoff were used.
Location of the Lesion
Localization of a lesion to a colonic segment is also a rough estimate with optical colonoscopy because the endoscopist has little direct extraluminal reference with which to correlate the position of the endoscope within the colon. Even with new methods, such as a magnetic device attached to the endoscope that indicates the location of the endoscope on the skin of the patient, optical colonoscopy is subject to overestimation of the distance of the lesion from the rectum because of the stretching of the colon by the endoscope. Although the comparison for lesion location is best done by using localization of the end of the endoscope with fluoroscopy and possibly by reviewing a recorded video of the endoscopic examination, it is impractical to require this of every endoscopist, and the value of video review is unproved.
Thus, the matching of lesions between optical colonoscopy and CT colonography should be done jointly by reviewing the CT colonography and endoscopy reports and images. Because many trials forbid unblinding of the researchers until the end of the trial, we suggest that investigators further document the location of the lesion at endoscopy by determining the distance from the anal verge and by acquiring still images of each lesion. Although shallow folds may efface with maximal distention, the relationship of a lesion to thick folds (eg, location of a polyp on a fold vs in between folds) and the morphologic structure of the lesion (sessile, pedunculate, or flat) can be helpful clues and should be reported. When multiple lesions are present, the sequence of lesions from the anal verge can also be helpful. Investigators should state clearly what methods they employed to compare the size and position of polyps.
When reporting the location of a lesion, it is still desirable to estimate segmental location. The reason is that the segmental distribution of lesions is both an intuitively understood metric and one that permits easy analysis of the anatomic distribution and prevalence of different lesions. In the past, schemes employing six or eight segments have been used. We suggest avoiding "hepatic flexure" and "splenic flexure" as separate segments. Although the location of the flexures is usually well defined for the radiologist, there is no corresponding clear transition point between the flexures and the adjacent colonic segments for the endoscopist.
Therefore, we suggest the use of rectum, sigmoid colon, descending colon, transverse colon, ascending colon, and cecum as segments to report the location of a lesion. The rectum includes the dentate line to the most proximal valve of Houston. The sigmoid colon extends from the most proximal valve of Houston to the inferior aspect of the vertical segment of descending colon. The latter is sometimes a poorly defined point at CT colonography; however, the descending colon normally occupies a clearly retroperitoneal position, which is identifiable at CT. The most cephalic curvatures of the flexures define the proximal and distal edges of the transverse colon. The ascending colon continues inferiorly from the proximal boundary of the transverse colon until the ileocecal valve, which defines the border between the cecum and the ascending colon. This last boundary can be established by drawing a line from the valve perpendicular to the centerline of the colon.
It is also important for investigators to provide histologic data, when available, because for average-risk patients, carcinomas appear to arise from benign adenomas (12). Hyperplastic polyps are generally not considered premalignant, although there are emerging reports that hyperplastic polyps (especially when located proximally) may be a precursor of malignancy through a rapidly progressive mutator pathway (13). At the time of this writing, this pathway is not clearly understood. Several investigators have suggested that hyperplastic polyps may flatten and may be less conspicuous in well-distended or overdistended colons (14,15). We suggest that histologic data should be reported for all lesions, including hyperplastic polyps, adenomatous polyps, and carcinomas. Unless histologic findings and the morphologic structure of the lesion are the focus of the research, it is impractical to require more detailed reporting of morphologic structure for lesions with villous or dysplastic histologic findings. We recommend reporting by-patient sensitivity and specificity data for adenomatous lesions separately, in addition to a combined statistic for polyps of all histologic types, to understand the importance of the prevalence of hyperplasic polyps in the cohort studied.
Synchronous Lesions
It is possible that the presence of multiple synchronous lesions could affect performance at CT colonography for polyp detection. For example, one might argue that the observation of even a single lesion that is 5 mm or 10 mm at CT colonography will result in a colonoscopic follow-up examination. Hence, to decrease reading time and reader fatigue, a detailed search for synchronous lesions might not be undertaken if a lesion has already been detected.
However, because optical colonoscopy may be incomplete in 2%–10% of patients and because the actual "miss rate" for optical colonoscopy is estimated at 6% for 10-mm lesions (9), we suggest that CT colonographic interpretation should involve a complete evaluation of the entire colon. Alternately, it is possible that the detection of a first lesion might increase the likelihood of detecting nearby synchronous lesions by increasing the time and attention spent by the radiologist in evaluating the affected segment of colon; if one spends more time evaluating a polyp candidate, there is a greater likelihood of detecting a second adjacent lesion when one is present. However, as the number of synchronous lesions increases, there may be a greater likelihood that some will be overlooked. To our knowledge, the magnitude of these effects is currently unknown and will be elucidated only with further data. Because of these complex biases that could affect sensitivity, the number of patients with synchronous lesions, the lesion size, and the histologic findings in those patients should be specified.
Morphology of the Lesion
When feasible, polyps should be reported with a morphologic description because different morphologic structures are thought to confer different risks of harboring carcinoma (16). For the purposes of consistency and simplicity, we suggest the following standard descriptions: pedunculate, sessile, and flat. Pedunculate lesions are round or mushroom-cap shaped and can demonstrate a distinct stalk of attachment to the colonic mucosa. Sessile lesions are broader mesa-shaped lesions with a base of attachment to the wall of the colon that is greater in width than the height of vertical encroachment of the lesion into the lumen of the colon. Flat lesions have only recently been described in the literature and are the least well understood in terms of their reported prevalence and clinical relevance.
A great deal of attention has been paid recently to the performance of CT colonography for detecting flat lesions (17,18). At the time of this writing, there is a paucity of data to clarify the clinical importance of these lesions or their prevalence in different populations. The findings in prior reports in the pathology and endoscopy literature have demonstrated a high degree of dysplasia in small flat lesions (19,20), which suggests that these lesions may not follow the usual time course postulated for the adenoma-carcinoma sequence (12). In addition, it has been suggested that the conspicuity of flat lesions at CT colonography may be compromised, depending on the limited projection of these lesions into the lumen of the colon (17). However, recently, Fidler et al (17,18) have shown that a flat lesion, defined as a lesion with a diameter that is twice its height, is less common than previously reported and that these lesions were also often hyperplastic, rather than adenomatous or dysplastic. These new data appear to contradict previous pathology reports.
We propose that for reporting purposes at CT colonography, a flat lesion should be defined as one that is 2 mm or less in height (19,20) with respect to the adjacent normal mucosa. (Although the best definition is a histologic one, requiring the flat lesion to be no more that two times the height of the mucosal thickness, this is an impractical definition for CT colonography [19,20].) Accordingly, when a patient is suspected of having a flat lesion, we cannot require but do strongly recommend that the additional dimension of the height of the lesion as shown at CT colonography (and, when possible, at optical colonoscopy) should be reported. Although the height of the lesion may be measured easily from the CT colonographic images, we recognize that lesion height determined from endoscopic examination may be only an estimate and is not normally reported unless prospectively requested.
Because lesions that project into the lumen are more amenable to analysis with current computer-aided detection programs, compared with infiltrative lesions that do not project into the lumen (21), we suggest that investigators who focus on flat lesions consider further subdividing them into (a) those that are infiltrative, with no perceptible raised component, and (b) those that are raised and project at least a few millimeters into the lumen. Optionally, so-called carpet lesions and other surface textural features, such as smooth versus lobulate morphologic structure, may be reported.
Definition of the "Gold Standard"
Clearly, optical colonoscopy is our best arbitrary "standard," but it is important that we recognize its limitations, particularly its false-negative rate, and the strategies for segmental unblinding, as discussed previously. In addition, when possible, all follow-up data should be used for comparison of lesion matching, including surgical findings and follow-up endoscopic results.
If a follow-up examination (a second colonoscopy, flexible sigmoidoscopy, barium enema study, or surgical resection of colon) changes the "truth," then the by-patient and by-polyp data should be presented both with and without these follow-up data. This can be limited to a practical period of time, such as 3 months, because colonic polyps are thought to grow slowly. When possible, investigators should always include a retrospective analysis of the cause of false-positive and false-negative interpretations of CT colonographic and optical colonoscopic findings.
Radiation Dose
The radiation dose associated with CT colonography depends on the underlying imaging technique and is a subject of active investigation at the time of this writing. The introduction of multidetector technology has permitted a decrease in the section thickness that may be used for complete imaging of the colon in a single breath hold. The use of thinner sections has the advantage of better z-axis resolution, resulting in nearly isotropic voxels and improved multiplanar reconstructions. These views facilitate reading colonic segments that are more readily observable in the coronal or sagittal plane (such as the transverse and sigmoid colon) because of their orientation orthogonal or oblique to the direction of table travel. Improved z-axis resolution also results in improved three-dimensional reconstructions and may improve the accuracy of lesion size measurement and morphologic analysis. However, to some degree, thinner sections are associated with higher noise and require an increase in tube current to maintain image quality.
Images for CT colonography may be acquired with substantially decreased dose, exploiting the fact that the high soft-tissue contrast that exists for a polyp surrounded by the gas of the insufflated colon reduces the dose required to image the polyp adequately. Although decreasing x-ray tube current sacrifices the characterization of extracolonic soft-tissue lesions, preliminary data suggest that substantial dose reduction is possible before performance for detection of polyps is affected (22,23). To date, several investigators have reported adequate performance for detection of polyps by using doses that ranged from 20% to 50% of that used for standard adult diagnostic CT technique, for example, 35–50 mAs effective, 140 kVp. It remains unclear, however, how further attempts at dose reduction will be balanced with requirements to interpret extracolonic findings.
Several centers, led by the European researchers in particular, have advocated a reduction in the radiation dose with CT colonography to as low as reasonably achievable (24,25). Because consensus does not yet exist as to what constitutes "low" dose, investigators are encouraged to specify completely the technique and parameters employed for transverse image acquisition.
At this time, it is impossible to mandate a single method of reporting dose. To facilitate comparison of reports from different institutions, the parameters necessary to estimate dose should be given. These parameters include specifying the vendor and model of CT scanner and the CT technique (collimation; pitch or pitch mode, depending on vendor; table speed; tube rotation time; milliamperes; dose setting; reconstruction interval and algorithm).
When reporting radiation dose, two parameters are of interest. The "dose-length product" (in milligray-centimeters) is a generally accepted metric that describes the absorbed radiation dose for a given scan series and may be summed for all of the scan series in a single examination. Most vendors provide an estimate of the dose-length product for each scan series. In addition, an estimate of the "effective dose" (in millisieverts) to the patient is useful to permit comparison of the radiation dose to other imaging modalities. The effective radiation may be estimated with various methods. One relatively simple method described by Jessen et al (26) permits estimation of the effective dose from the dose-length product and is used in the CT Accreditation Program for the American College of Radiology.
Method of Interpretation
To compare studies and to pool results, precise reading conditions for observers should be specified (27). Although there is no definitive proof that reading platform or methods significantly (P < .05) affect diagnostic accuracy (28), they do affect interpretation time and are still being investigated (27). For this reason, software and hardware platforms should be specified by vendor, version, and other factors that can affect reading time and possibly diagnostic accuracy, such as single versus dual monitor reading, mouse versus button-control navigation, and automated versus manual fly-though of endoluminal reconstructions. Little head-to-head comparison has been made of these different viewing factors, and complete reporting of them will assist in retrospective analysis of their effects (27,29). Because cost analysis of CT colonography will be an important aspect of its evaluation, it is useful to separate interpretation time (given as a range with median or as an average with SD) from the time needed to fill out research case report forms.
Emphasis has been placed on the need for experience in interpreting the results of CT colonographic examinations. The qualifications of the readers should be defined in some way, such as the number of prior examinations read with colonoscopic or surgical proof. If there are other factors that might affect the accuracy of the reader, they should be specified (eg, pressure to give an immediate interpretation for same-day colonoscopy, or reading a large number of cases in a controlled setting in 1 day when fatigue may be a factor).
Quality Assurance and Confidence Scales
The number of nondiagnostic studies should be reported, as well as an assessment of how these studies affected performance data. Many investigators have used confidence scales to help in analyzing subjective data, and this is another source of variation among reports. For investigators who choose to use confidence scales, we suggest that rating scales for (a) quality of distention, (b) residual fluid, or (c) residual solid stool should use a five-point scale with discrete verbal descriptors of each of the points. Similarly, for evaluation of subjective reader confidence regarding the presence or absence of a polyp, the scale used should be well defined, with five or more points (or a continuous scale).
When multiple readers are employed, investigators should clarify if their interpretations are independent or grouped in consensus. Independent, multiple-observer studies are preferred.
By-Patient Data
By using the size and histologic criteria outlined previously, the by-patient sensitivity, specificity, and positive and negative predictive values of CT colonography should be reported with their CI and P values. Inclusion criteria often mix American Cancer Society average-risk and above-average-risk patients. The data should be reported both combined and, when appropriate, stratified by screening and diagnostic examination cohorts as defined by the criteria of the American Cancer Society (6).
Future Considerations
The importance of current research on stool opacification, electronic subtraction of fluid and stool, and integration of computer-aided diagnosis is uncertain, but advances in these techniques is likely to improve patient acceptance and ease of interpretation (30). In evaluating stool opacification, the dosing regimen should be detailed. For electronic subtraction, the algorithm for detection, subtraction, and mucosal reconstruction should be specified (31). In comparing computer-aided detection to observer studies, we suggest that computer-aided detection should be evaluated as a "second opinion," similar to its use at mammography.
In the long term, we must develop guidelines to link CT colonographic results to recommendations for patient care. CT colonography will force us to address the question: what is the appropriate management for lesions in the size categories listed previously? We must develop standard recommended intervals between examinations that take patient risk stratification and examination performance data into consideration. This last question falls into the category of decision analysis, an essential aspect of colonography that merits further investigation. In collaboration with our colleagues in gastroenterology, we need to determine what threshold of size constitutes a clinically relevant lesion for reporting and resection, and we need to collaborate in the development of an approach to the description and identification of lesions. All of these questions are essential ones that will be more readily answered with the largest and most consistently reported pool of data that we, as a community of investigators, can generate.
ACKNOWLEDGMENTS
The authors thank Jim Brink, MD, David T. Rubin, MD, and Dezheng Huo, MS, for their advice.
FOOTNOTES
A.H.D. is a consultant to E-Z-Em and GE Medical Systems and has research funding from E-Z-Em.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. van Wijk, J. Florie, C. Y. Nio, E. Dekker, A. H. de Vries, H. W. Venema, L. J. van Vliet, J. Stoker, and F. M. Vos Protrusion Method for Automated Estimation of Polyp Size on CT Colonography Am. J. Roentgenol., May 1, 2008; 190(5): 1279 - 1285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Robinson, S. Halligan, S. A. Taylor, S. Mallett, and D. G. Altman CT Colonography: A Systematic Review of Standard of Reporting for Studies of Computer-aided Detection Radiology, February 1, 2008; 246(2): 426 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Kim, J. M. Lee, H. W. Eun, M. W. Lee, J. K. Han, J. Y. Lee, and B. I. Choi Two- versus Three-dimensional Colon Evaluation with Recently Developed Virtual Dissection Software for CT Colonography Radiology, September 1, 2007; 244(3): 852 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Park, E. K. Choi, S. S. Lee, J.-S. Byeon, J.-Y. Jo, Y. H. Kim, K. H. Lee, H. K. Ha, and J. K. Han Polyp Measurement Reliability, Accuracy, and Discrepancy: Optical Colonoscopy versus CT Colonography with Pig Colonic Specimens Radiology, July 1, 2007; 244(1): 157 - 164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Park, S. S. Lee, E. K. Choi, S. Y. Kim, S.-K. Yang, J. H. Kim, and H. K. Ha Flat Colorectal Neoplasms: Definition, Importance, and Visualization on CT Colonography Am. J. Roentgenol., April 1, 2007; 188(4): 953 - 959. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Pickhardt, V. T. Lehman, T. C. Winter, and A. J. Taylor Polyp volume versus linear size measurements at CT colonography: implications for noninvasive surveillance of unresected colorectal lesions. Am. J. Roentgenol., June 1, 2006; 186(6): 1605 - 1610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Halligan, S. H. Park, and H. K. Ha Causes of False-Negative Findings at CT Colonography Radiology, March 1, 2006; 238(3): 1075 - 1077. [Full Text] [PDF]
|
|
|
|
|
|
T. Yasumoto, T. Murakami, H. Yamamoto, M. Hori, R. Iannaccone, T. Kim, H. Abe, M. Kuwabara, K. Yamasaki, N. Kikkawa, et al. Assessment of Two 3D MDCT Colonography Protocols for Observation of Colorectal Polyps Am. J. Roentgenol., January 1, 2006; 186(1): 85 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Halligan, D. G. Altman, S. A. Taylor, S. Mallett, J. J. Deeks, C. I. Bartram, and W. Atkin CT Colonography in the Detection of Colorectal Polyps and Cancer: Systematic Review, Meta-Analysis, and Proposed Minimum Data Set for Study Level Reporting Radiology, December 1, 2005; 237(3): 893 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Iannaccone, C. Catalano, F. Mangiapane, T. Murakami, A. Lamazza, E. Fiori, A. Schillaci, D. Marin, I. Nofroni, M. Hori, et al. Colorectal Polyps: Detection with Low-Dose Multi-Detector Row Helical CT Colonography versus Two Sequential Colonoscopies Radiology, December 1, 2005; 237(3): 927 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Pickhardt, A. D. Lee, E. G. McFarland, and A. J. Taylor Linear Polyp Measurement at CT Colonography: In Vitro and in Vivo Comparison of Two-dimensional and Three-dimensional Displays Radiology, September 1, 2005; 236(3): 872 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Zalis, M. A. Barish, J. R. Choi, A. H. Dachman, H. M. Fenlon, J. T. Ferrucci, S. N. Glick, A. Laghi, M. Macari, E. G. McFarland, et al. CT Colonography Reporting and Data System: A Consensus Proposal Radiology, July 1, 2005; 236(1): 3 - 9. [Full Text] [PDF]
|
|
|
|
|
|
M. E. Zalis, J. J. Perumpillichira, J. Y. Kim, C. Del Frate, C. Magee, and P. F. Hahn Polyp Size at CT Colonography after Electronic Subtraction Cleansing in an Anthropomorphic Colon Phantom Radiology, July 1, 2005; 236(1): 118 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lefere, S. Gryspeerdt, J. Marrannes, M. Baekelandt, and B. Van Holsbeeck CT Colonography After Fecal Tagging with a Reduced Cathartic Cleansing and a Reduced Volume of Barium Am. J. Roentgenol., June 1, 2005; 184(6): 1836 - 1842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Park, H. K. Ha, M.-J. Kim, K. W. Kim, A. Y. Kim, D. H. Yang, M.-G. Lee, P. N. Kim, Y. M. Shin, S.-K. Yang, et al. False-Negative Results at Multi-Detector Row CT Colonography: Multivariate Analysis of Causes for Missed Lesions Radiology, May 1, 2005; 235(2): 495 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Ferrucci Colonoscopy: Virtual and Optical--Another Look, Another View Radiology, April 1, 2005; 235(1): 13 - 16. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ferrucci, M. Barish, R. Choi, A. Dachman, H. Fenlon, S. Glick, A. Laghi, M. Macari, M. Morrin, E. Paulson, et al. Virtual Colonoscopy JAMA, July 28, 2004; 292(4): 431 - 432. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
