| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Departments of Radiology (R.E.v.G., H.W.V., J.F., C.Y.N., M.P.S., F.M.V., J.S.L., J.S.), Medical Physics (H.W.V.), Clinical Epidemiology and Biostatistics (J.C.v.R., A.S.G., P.M.M.B.), and Gastroenterology (J.F.W.B.), Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands; and the Pattern Recognition Group, Department of Applied Physics, Technical University, Delft, the Netherlands (I.W.O.S., F.M.V.). Received July 8, 2003; revision requested September 16; revision received October 29; accepted December 15. Supported by Netherlands Organization for Health Research and Development grant 2100.0059. Address correspondence to R.E.v.G. (e-mail: r.e.vangelder@amc.uva.nl).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
© RSNA, 2004
Index terms: Colon, CT, 75.12115, 75.12117 • Colon neoplasms, CT, 75.12115, 75.12117 • Computed tomography (CT), radiation exposure, 75.12115, 75.12117
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Unfortunately, exposure of asymptomatic subjects to radiation doses ranging from 4 to 12 mSv, as are used routinely today (11), with an estimated risk of one fatal cancer in 3,300–10,000 individuals aged 50 years (12), remains a major drawback. If the screening test requires regular repetition, the risk increases proportionally with the number of repetitions.
In clinical practice, the benefits of CT scanning for the patients generally outweigh the drawback of use of ionizing radiation. This balance, however, is very delicate in patients with a low potential benefit, as in the screening of asymptomatic people. Consequently, use of the lowest achievable dose for CT colonography will improve this balance and may serve implementation of this technique in colorectal cancer screening. This goal is facilitated by the recent introduction of CT scanners that enable use of lower doses than have been possible previously, a development driven by the present awareness of the potential danger of ionizing radiation (13–15).
To date, several investigators have studied the feasibility of low-dose protocols for CT colonography; their findings demonstrate that sensitivity and specificity are still satisfactory (11,16–19), even with an effective dose down to 1 mSv (18). However, the dose at which polyp detectability starts to decrease is still unknown, as is the lowest dose that is compatible with acceptable polyp detection.
The aim of this feasibility study was to compare polyp detection and interobserver variability at CT colonography with four low to very low doses (down to 0.05 mSv) with those at medium-dose (8–12 mSv) CT colonography.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Patient Population
For this study, we included prone and supine CT scans obtained in 15 patients with at least one polyp 5 mm or larger in largest dimension who had participated in an accuracy study of CT colonography. The Medical Ethics Committee of the Academic Medical Center, Amsterdam, the Netherlands, approved this accuracy study, and all patients gave written informed consent. Because the current study consisted of modification of data that had already been collected and no additional CT scans were obtained, the Medical Ethics Committee indicated that no additional approval from them and no additional informed consent from the patients were required.
All 15 patients were referred from the Academic Medical Center to undergo colonoscopy because of surveillance (personal or family history of colorectal polyps or cancer) or polypectomy. Ten men and five women (mean age, 59 years ± 13 [standard deviation]) were enrolled in the study. Mean age for the men was 55 years ± 12 and for the women was 67 years ± 12. All subjects used 4 L of macrogol bowel preparation (KleanPrep; Helsinn Birex Pharmaceuticals, Dublin, Ireland) prior to the examinations.
CT Colonography
Each patient underwent imaging with a multisection CT scanner (Mx8000; Philips Medical Systems, Best, the Netherlands) in prone and supine positions. To distend the colon, approximately 2 L of air (13.4% volume of carbon dioxide) was insufflated manually. Before a patient underwent scanning, a preview image was obtained to estimate the distension; if necessary, additional air was insufflated with approval of the patient. CT was performed with the following parameters: 120 kV; collimation, 4.0 x 2.5 mm; rotation time, 0.75 second; pitch, 1.25; section thickness, 3.2 mm; and reconstruction interval, 1.6 mm. For all but one patient, 100 mAs was used. That patient was the last, who underwent CT after the overall milliampere-second level was reduced for CT colonography, and 65 mAs was used. The effective dose for a complete examination was 8 mSv (at 65 mAs) or 12 mSv (at 100 mAs). The effective dose was estimated as an average for men and women by using a CT dosimetry calculator (ImPACT; Medicines and Healthcare Products Regulatory Agency, London, England). The calculated effective dose applied only to average-sized patients. For the same CT scanning parameters, the effective dose in heavier patients would be somewhat lower and that in thinner persons would be somewhat higher. At present, we do not have the facilities to perform patient-specific effective dose calculations.
Colonoscopy
Within 2 hours after CT colonography, all patients underwent colonoscopy with a standard endoscope (CF-140L; Olympus, Tokyo, Japan). Four of the colonoscopic examinations were performed by three gastroenterologists (including J.F.W.B.), who each had experience of 5 years or longer. The other 11 colonoscopic examinations were performed by eight gastroenterology fellows (four in their 1st year, two in their 2nd year, and two in their 3rd year) with the direct supervision of the attending gastroenterologist (J.F.W.B.). The endoscopists were blinded to results at CT colonography. During this examination, 13 patients received midazolam hydrochloride (Dormicum; Roche, Basel, Switzerland) and 11 received fentanyl citrate (Fentanyl-Janssen; Janssen Pharmaceuticals, Beerse, Belgium). A research fellow (R.E.v.G., a medical doctor but not a radiologist) recorded the examination on videotape. Polyps, if present, were scored with respect to the size, morphology, and segment (cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum) in which they were located. Polyp size (largest dimension) was estimated by the endoscopist who performed colonoscopy on the basis of comparison with an open biopsy forceps of known size.
Dose Reduction at CT
Radiation-dose reduction at CT scanning can be accomplished by reducing the number of photons to which the patient is exposed. As a consequence, image noise is increased. The number of photons can be reduced by reducing the tube current, increasing the pitch, or both. With the multisection CT scanner used in the current study, the tube current is automatically adjusted to the selected pitch, as the quantity effective milliampere-second level (tube current multiplied by the quotient of rotation time divided by pitch) is used; as a consequence, both patient dose and image noise depend on only the chosen (effective) milliampere-second value (20). In this study, CT colonographic examinations were simulated with a lower milliampere-second value and thus with a lower dose.
Simulation of Low-Dose CT Colonography
To compare polyp detection and interobserver variability at CT colonography with medium and lower doses in the same 15 patients, we used an established simulation technique (11,21,22) to obtain four additional lower-dose CT colonographic images in each patient. Findings in a previous study (22) demonstrate that simulated lower-dose CT scans obtained with a similar method cannot be distinguished from genuine lower-dose scans.
The simulation method consists of controlled increase of noise in the colonoscopic raw CT transmission data prior to reconstruction of the images. The essence of the simulation method used in the current study is the following (21). First, the number of photons corresponding to each transmission measurement was determined by using calibration measurements with rectangular water-filled phantoms and calculations of the attenuation of the CT x-ray spectrum in water. Next, the number of detected photons corresponding to a lower milliampere-second value was simulated by generating a Poisson deviate with the appropriate lower mean value. Finally, this Poisson deviate was converted back into a transmission measurement that corresponded to the desired lower tube current time. In this simulation, the electronics noise floor and other noise sources apart from quantum noise were not taken into account. The procedure was validated with CT scans obtained at various milliampere-second levels in water-filled phantoms by comparing the standard deviations of the simulated and real CT images. For very low milliampere-second levels (less than 15 mAs, with which no scans can be obtained with the multisection CT scanner used in this study), the procedure was validated with CT scans obtained in water-filled phantoms by using Monte Carlo simulations in which the complete measurement and reconstruction process was taken into account.
In practice, a research fellow (R.E.v.G.) extracted the raw transmission data from the multisection CT scanner, increased the noise of the raw data off-line by using software written by the physicist (H.W.V.), reentered the modified raw data into the multisection CT scanner, and made reconstructions by using the standard reconstruction software. The resulting CT images simulate scans acquired with reduced milliampere-second level and thus with proportionally reduced dose. Scans were simulated at 25.0, 6.3, 1.6, and 0.4 mAs, which correspond to effective doses for prone and supine scans together of 3.0, 0.8, 0.2, and 0.05 mSv, respectively. The two lowest doses are not yet achievable in clinical practice, but they are technically possible. Reduction by a factor of four for consecutive milliampere-second levels has the consequence that, to a first approximation, the noise level is doubled in the corresponding CT scans.
Noise Reduction Filters for Very Low Milliampere-Second Level
Lowering of the milliampere-second level increases the image noise, if all other parameters are kept constant. For the lowest simulated milliampere-second values used in this study, this increase in noise was such that the two-dimensional images (Fig 1) and especially the three-dimensional images (Fig 2) had very poor quality. Therefore, smoothing of images at a simulated lower milliampere-second level was required to counteract this increase in noise. For reconstruction of the original CT data obtained with 100 and 65 mAs, a standard reconstruction filter was used (filter C of the multisection CT scanner). The simulated images obtained at lower milliampere-second levels were reconstructed with the smoothest reconstruction filter available (filter A of the multisection CT scanner). For the lowest three doses, additional in-plane smoothing of the CT images was used by convolving the images with a Gaussian kernel with 
= 0.6 mm (6.3 mAs) and = 1.2 mm (1.6 and 0.4 mAs). This amount of smoothing was chosen empirically to keep the noise in the images approximately constant with decreasing tube current times, down to the simulated 1.6-mAs images. To not reduce the spatial resolution any further, we used the same filters for the 0.4- and 1.6-mAs simulated images.
|
|
|
|
Ninety segments were examined; 33 (37%) of them contained at least one polyp. Of these 33 segments, 13 contained only polyps smaller than 5 mm and 20 contained polyps 5 mm or larger. Among the latter segments, five contained a polyp that was 10 mm or larger.
In the present study, CT colonographic data were evaluated with a primary three-dimensional display mode. Data from a previous study (23), as well as from our own preliminary experience, indicate that with primary three-dimensional display modes a better polyp visualization is achieved than that with primary two-dimensional displays. To achieve optimal colon surface visibility, we used three-dimensional unfolded cubic projections (24). This method comprises the evaluation of a sequence of preprocessed three-dimensional unfolded cubic projections of the prone and supine CT data with the ability to further investigate suspected lesions with two-dimensional images. All CT colonographic data were volume rendered (threshold, –750 HU; ramp, 80 HU). The observers measured the lesions they detected in the two-dimensional images because, with the software we used, this was more time efficient than measurement in three-dimensional images. The evaluation was performed with a workstation (EasyVision; Philips Medical Systems).
Definitions of True- and False-Positive Findings
Another research fellow (R.E.v.G.) compared all findings at CT colonography with those at colonoscopy, which was the reference standard, by means of direct visual comparison of the recorded CT colonographic images and the videotaped colonoscopic images. This research fellow had previously matched CT colonographic and colonoscopic findings in more than 100 patients but was not involved in the review of CT colonographic images (performed by C.Y.N. and J.F.). A polyp detected at CT colonography was labeled as a true-positive finding if it corresponded exactly to a polyp detected at colonoscopy in terms of location (segment: cecum; ascending, transverse, descending, and sigmoid colon; and rectum), morphology, and size. Because estimates of the size of polyps at colonoscopy tend to have errors, we accepted a margin of error of 3 mm for polyps smaller than 5 mm and a margin of 5 mm for polyps 5 mm or larger. All other findings at CT colonography were classified as false-positive.
Statistical Analysis
Detection parameters.—Polyp sensitivity at each dose was defined as the number of true-positive findings relative to the total number of polyps.
Because polyp specificity cannot be defined, we report the number of false-positive findings and the segment specificity, defined as the number of segments without polyps at CT colonography relative to the number of segments without polyps at colonoscopy.
Although some investigators advocate the removal only of polyps 10 mm or larger (25), clinicians and patients are unlikely to accept the fact that lesions between 5 and 10 mm remain undetected (26,27). Therefore, we used 5 mm as the main cutoff in the present study. Because this matter is a subject of continuous discussion, however, we also used 10 mm as a cutoff in the analysis of true- and false-positive findings.
The true-positive findings were subdivided on the basis of the size estimated by the endoscopists, and the false-positive findings were subdivided on the basis of the measurements made by each observer on the CT images.
Stratification of polyp sensitivity according to pelvic and abdominal location.—In the pelvic region, the noise level is generally higher than that in other parts of the abdomen because of the larger lateral dimensions of the pelvis and the presence of bone structures. The effect of dose reduction on image quality will therefore be most severe in this region. For this reason, we performed an additional analysis to study the effect of dose reduction on sensitivity for polyps 5 mm or larger in segments in the pelvis (cecum, sigmoid colon, and rectum) and those in the abdomen (ascending, transverse, and descending colon).
Differences in polyp detection between the colonoscopic and lower-dose CT colonographic data were assessed with the McNemar test for paired proportions. Because few polyps were 10 mm or larger, differences in detection of these polyps between colonoscopic and lower-dose scans could not be assessed. Differences in polyp sensitivity between the pelvic and abdominal regions were assessed at each dose with the Mann-Whitney test. All analyses were performed separately for each observer.
Interobserver agreement.—Interobserver agreement analysis was performed on the basis of segment for each dose. Observers were considered to agree if they recorded one or more polyps in the same segment, regardless of whether these findings were true- or false-positive. If both observers recorded no findings in the same segment, they were also considered to agree. The 
statistic with 95% confidence interval was calculated. The values were interpreted on the following basis: < 0.20, poor agreement; = 0.21–0.40, fair agreement; = 0.41–0.60, moderate agreement; = 0.61–0.80, good agreement; = 0.81–1.00, very good agreement.
Interpretation time.—The time required to interpret findings in each segment was registered automatically with the workstation. Interpretation times for segments in the same patient were added to obtain patient interpretation times. Differences in interpretation times between the colonoscopic and lower-dose CT colonographic images were assessed with the paired-samples t test for each observer separately.
Differences with a P value less than .05 were considered significant. Software (SPSS for Windows, version 11.0; SPSS, Chicago, Ill) was used for all statistical tests.
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
CT Colonography
True- and false-positive lesions.—True- and false-positive findings at CT colonography for all milliampere-second levels are listed in Table 1. Detection of polyps of all sizes decreased significantly (P < .05) with decreasing milliampere-second level, except at 25 mAs for observer 1. Down to 1.6 mAs, the mean sensitivity for polyps 5 mm or larger decreased slightly but remained equal to or higher than 74%, as shown in Figure 5. For observer 1, sensitivity for polyps 5 mm or larger varied from 85% to 78%, whereas for observer 2 this parameter varied from 81% to 70%. At the lowest milliampere-second level, sensitivity for polyps 5 mm or larger decreased to 44% for observer 1 and to 52% for observer 2. None of the differences in detection of polyps 5 mm or larger between the 100 mAs and simulated lower milliampere-second data were significant, except between 100 and 0.4 mAs (P < .05), for both observers. Detection of large polyps was between 80% and 100% at all milliampere-second levels.
|
|
Segment specificity.—Table 2 shows the segment specificity. Specificity regarding the correct identification of segments without polyps increased or remained constant, except at 0.4 mAs. Concerning segments without polyps 5 mm or larger, specificity improved for observer 1 from 79% at 100 mAs to 93% at 0.4 mAs and for observer 2 was between 90% and 96% at all milliampere-second levels. Specificity for segments without polyps 10 mm or larger was between 96% and 99% at all milliampere-second levels for both observers.
|
|
Interobserver analysis.—Results of interobserver analysis are given in Table 4. The statistics for CT colonography were good at all milliampere-second levels, except for the lowest value for which the statistic was moderate.
|
|
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
The influence of dose on the detectability of polyps depends on their size, which therefore will play an important role in the determination of radiation dose. A significant decrease in sensitivity for polyps 5 mm or larger was found only at 0.05 mSv, whereas sensitivity for polyps 10 mm or larger did not decrease. Owing to the small number of polyps 10 mm or larger, however, we were not able to statistically assess the latter observation.
In the present feasibility study, we aimed to find the order of magnitude of the lowest dose that is compatible with acceptable polyp detection. To guarantee internal validity, we applied five doses in 15 patients to make sure that the differences in polyp visibility were purely a result of the differences in dose. A potential disadvantage of this setup is that memory bias may have been present. We applied three strategies to reduce this potential bias. First, we employed an interval of at least 4 weeks between two subsequent readings. Second, data for each segment were evaluated separately and in a random order to further reduce memory bias. Finally, the readings started with the lowest dose followed by successively higher doses. Because of the large difference between doses, it is extremely improbable that, even if the observer still had any memory of the previous reading, this information would assist interpretation of the new version with a lower noise level.
This study design has some disadvantages, particularly regarding the possibility of generalizing these findings to a clinical setting. Because the potential future role of CT colonography is in the selection of patients with at least one polyp of a certain size but not in the identification of each polyp, our outcome parameters do not directly apply to clinical practice. Also, because we primarily aimed to find the lowest dose compatible with acceptable polyp detection, a prerequisite for potential implementation of CT colonography, we confined our population to patients with polyps, and consequently patient specificity could not be calculated. It is conceivable, however, that the relative effect of dose reduction on segment specificity in the present study is comparable to that on patient specificity. Future studies, performed in a consecutive surveillance or screening population, that also include patients without polyps are required to establish the clinical value of very low–dose CT colonography in terms of patient sensitivity, specificity, and predictive values.
The number of false-positive findings decreased with decreasing dose in this study. The increased noise and the smoothing that was applied in the images obtained with the three lowest doses may have restrained observers from considering small irregularities as polyps.
In this study, both observers had a high number of false-positive findings, especially at 100-mAs and simulated 25-mAs CT colonography. This result appears to be at least partially caused by the unexpectedly high number of segments that were contaminated with fecal residue. Most false-positive lesions were smaller than 5 mm, but one observer detected a considerable number of false-positive lesions that were 5 mm or larger. Because these false-positive findings were confined to 15 of 70 segments without polyps 5 mm or larger, the segment specificity was still considerable (79%).
Although segment specificity for both readers was similar, the experienced abdominal radiologist had a much higher number of false-positive findings than did the research fellow. Because both observers had comparable experience in CT colonography, the cause of this discrepancy is not clear.
Pelvic polyp sensitivity was slightly lower than abdominal polyp sensitivity. This difference may be a result of the higher noise level in the pelvic region and the more difficult interpretation of lesions in the tortuous sigmoid colon. This difference was significant at the lowest dose. Hence, dose reduction may be more complicated in the pelvis. The consequences of this finding must be further elucidated, especially in view of the possibilities of state-of-the-art CT scanners to adjust the tube current to the scan region. For constant overall image quality, tube current must be adjusted to the girth of the patient (11,28).
We adjusted the image smoothing to the dose, as would be done in clinical practice. After smoothing, the 6.3- and 1.6-mAs images had a noise level similar to that in the 25-mAs images, at the expense of spatial resolution. The in-plane resolution of CT images is usually adjusted by the choice of the reconstruction filter. The amount of smoothing required could not be obtained with the reconstruction filters available on the multisection CT scanner. Therefore, additional filtering was obtained with spatial-domain filtering of the CT scans. Equivalent results have been obtained with either of these methods (29,30).
The lowest tube current times that were tested in this study (1.6 and 0.4 mAs) cannot yet be realized in clinical practice. Until recently, the possibility of CT scanning with very low doses was hampered by the presence of an electronics noise floor in the preamplifiers of the CT detectors, which increased the noise in the CT scans disproportionally at the lowest doses (31). CT scanners that were introduced during the past year profit from the development of improved detector technology, and scans can be acquired at a minimum of 3–5 mAs. Further reduction of milliampere-second level is technically possible and, as findings in the present study indicate, may be clinically useful.
The results of our study concur with those in other studies and extend them to substantially lower doses. Cohnen and colleagues (18) compared sensitivity and specificity with medium and very low–dose scan protocols (12 and 1 mSv, respectively) in two groups of patients suspected of having colorectal polyps or cancer. They found that the performance for polyps 5 mm or larger was similar: 89% (eight of nine polyps) and 94% (17 of 18 polyps) with the medium and very low doses, respectively. Iannaccone and colleagues (19) determined the accuracy for CT colonography performed with a 2-mSv scan protocol. They found excellent sensitivity and specificity, with detection of six of six polyps larger than 10 mm, nine of nine colorectal cancers, and no false-positive findings. These data are similar to previously reported results about the accuracy of CT colonography. A drawback of both studies is that the investigators did not compare different doses in identical patients, which means that other factors could account for the similar sensitivity and specificity that were observed. Our approach enables such a comparison without the drawbacks of acquisition of CT scans in the same patients at several doses and thus exposure to additional ionizing radiation.
Rigorous reduction of radiation dose may hamper the potential benefit of CT colonography to depict extracolonic findings because the image quality of solid organs is substantially deteriorated. However, the primary aim of CT colonography is the selection of patients with polyps, and therefore we did not take extracolonic findings into account.
The burdensome bowel preparation that was used in the present study may be redundant in the near future. Results of recent studies (32,33) show that the addition of contrast agents to several low-fiber diets enables distinction of bowel wall from bowel contents. In another study (34), intravenous contrast media were used to improve the conspicuity of the bowel wall and medium-sized polyps. The use of contrast agents may be incompatible with very low–dose CT because of the relatively low contrast between contrast material–enhanced and unenhanced structures.
Although CT colonography holds promise for colorectal cancer screening, the prospect of exposure of large numbers of asymptomatic individuals to intermediate doses of ionizing radiation, probably more than once in their lives, may hamper implementation in screening programs.
Exposure to ionizing radiation may induce cancer in the exposed individual after a latent period of up to a few decades. The risks of effective doses in the magnitude presently employed for CT colonography (4–12 mSv) remain uncertain (35) but cannot be disregarded, and therefore it is generally agreed that radiologists should aim for the lowest dose required for the diagnostic task.
In the case of CT scans obtained with an effective dose of approximately 0.2 mSv, which may be feasible on the basis of findings in the present study, the problems with exposure to ionizing radiation may be substantially reduced. Compared with the radiation doses of 4–12 mSv that are used routinely today (11), future doses would be reduced 20- to 60-fold.
Generalizability of findings in the present study may be limited because we employed a primary three-dimensional display mode as opposed to the primary two-dimensional display mode used by the majority of researchers. Although the two-dimensional images were less affected by the dose reduction than were the three-dimensional images, the effects of dose reduction on primary two-dimensional performance remain to be determined.
We conclude that findings in this feasibility study indicate that the effective dose in CT colonography can probably be considerably decreased without substantial impairment of detection of polyps 5 mm or larger and interobserver agreement. Future studies are needed to investigate the clinical value of CT colonography at a very low dose in a cohort of consecutive surveillance or screening patients.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. H. de Vries, H. W. Venema, J. Florie, C. Y. Nio, and J. Stoker Influence of Tagged Fecal Material on Detectability of Colorectal Polyps at CT: Phantom Study Am. J. Roentgenol., October 1, 2008; 191(4): W181 - W189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Johnson, J. G. Fletcher, R. L. MacCarty, J. N. Mandrekar, W. S. Harmsen, P. J. Limburg, and L. A. Wilson Effect of Slice Thickness and Primary 2D Versus 3D Virtual Dissection on Colorectal Lesion Detection at CT Colonography in 452 Asymptomatic Adults Am. J. Roentgenol., September 1, 2007; 189(3): 672 - 680. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Lucey, J. C. Varghese, A. Hochberg, M. A. Blake, and J. A. Soto CT-Guided Intervention with Low Radiation Dose: Feasibility and Experience Am. J. Roentgenol., May 1, 2007; 188(5): 1187 - 1194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Florie, S. Jensch, R. A. J. Nievelstein, J. F. Bartelsman, L. C. Baak, R. E. van Gelder, B. Haberkorn, A. van Randen, M. M. van der Ham, P. Snel, et al. MR Colonography with Limited Bowel Preparation Compared with Optical Colonoscopy in Patients at Increased Risk for Colorectal Cancer Radiology, April 1, 2007; 243(1): 122 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. WERDERMAN CT Colonography Radiol. Technol., March 1, 2007; 78(4): 309CT - 324CT. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Jensch, R. E. van Gelder, J. Florie, M. A. T.-d. Graaf, J. V. Lobe, P. M. M. Bossuyt, S. Bipat, C. Y. Nio, and J. Stoker Performance of Radiographers in the Evaluation of CT Colonographic Images Am. J. Roentgenol., March 1, 2007; 188(3): W249 - W255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mang, A. Maier, C. Plank, C. Mueller-Mang, C. Herold, and W. Schima Pitfalls in Multi-Detector Row CT Colonography: A Systematic Approach RadioGraphics, March 1, 2007; 27(2): 431 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Graser, B. J. Wintersperger, C. Suess, M. F. Reiser, and C. R. Becker Dose reduction and image quality in MDCT colonography using tube current modulation. Am. J. Roentgenol., September 1, 2006; 187(3): 695 - 701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Macari and E. J. Bini CT Colonography: Where Have We Been and Where Are We Going? Radiology, December 1, 2005; 237(3): 819 - 833. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP |
, 12 polyps) and pelvic (
, 15 polyps) regions separately. Overall sensitivity decreases slightly but remains 74% or more to 1.6 mAs and then decreases more sharply. Detection of abdominal and pelvic polyps slightly decreases. Only significant difference in pelvic polyp detection was between 100 and 0.4 mAs (P < .05).