| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, 1A71 School of Medicine, University of Utah School of
Medicine, 50 N. Medical Dr., Salt Lake City, UT 84132.
2
Department of Radiological Sciences, B2-247 CHS, UCLA School of Medicine,
10833 Le Conte Ave., Los Angeles, CA 90095.
3
Department of Radiological Sciences and Biomathematics, B2-200 CHS, UCLA
School of Medicine, Los Angeles, CA 90095.
4
Department of Cardiology, UCLA School of Medicine, Los Angeles, CA
90095.
Received June 15, 1998;
accepted after revision August 11, 1999.
Address correspondence to H.-C. Yoon.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One thousand patients (average age, 53 years; age range, 18-85 years) who were asymptomatic for coronary artery disease underwent two consecutive scans of the heart on an electron beam CT scanner. Scans were performed with ECG gating, breath-hold, 3-mm collimation, and 100-msec exposure. Two contiguous pixels with density values greater than 130 H were used as the minimum criterion for a calcific lesion. The calcium score was determined on a vessel-by-vessel basis for both scans of each patient. Interscan variation in calcium and vessels involved with calcification was evaluated on the basis of age, sex, and average calcium score.
RESULTS. The percentage of difference between calcium scores in scans was 28.4% and 43.0% for women and men, respectively. For the individual epicardial arteries (left main, left anterior descending, circumflex, and right coronary), the percentage of difference for calcium scores was 20.2-24.2% for women and 30.5-44.9% for men. A difference between the two scans in at least one vessel of the total coronary arteries identified with calcium was noted in 31% of patients.
CONCLUSION. Interscan variability in calcium scores may be important in the determination of risk stratification. Subjects with a nonzero calcium score may benefit from undergoing two scans at the time of initial imaging.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The coronary artery calcium score is measured as a quantitative score for each vessel and as a total score for all the epicardial arteries using a standard algorithm [6]. The number of calcific lesions and the number of vessels involved with calcification are also tabulated and may be reported. Both intraobserver and interobserver variation in calcium scores for a single scan are extremely low, with reliability coefficients greater than 0.99 [7], suggesting that only a single score need be generated for a scan. However, several studies have shown that calcium scores can have considerable interscan variation when two scans are performed on the same patient in close temporal proximity [8,9,10,11], and that this variability is more pronounced for patients with low but nonzero scores than for patients with high scores [11]. The aims of our study were to determine the effect of calcium distribution on score variations and to explore the potential clinical impact of interscan coronary artery calcium variation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of the 100 consecutive patients, 32 were found to have technical differences in their two scans or incomplete data on the risk factor questionnaires and were excluded from further analysis. Incomplete data on the risk factor questionnaires occurred in a number of patients who were imaged early in our study. Technologists did not sufficiently emphasize to our earliest patients the importance of obtaining complete risk factor data. The most common technical error observed in this patient sample was incomplete imaging of the entire heart on the first scan. Other less frequent sources of technical error included excessive patient motion on one of the two scans and poor cardiac gating. An additional 17 patients were found to have a history of heart disease discovered from the questionnaire, despite denying any disease or symptoms at the time the scans were scheduled. A total of 951 patients had two consecutive good-quality scans and appropriate cardiac risk factor data. The patients were 664 men (age range, 18-84 years; median age, 50 years) with a median of two coronary artery disease risk factors, and 287 women (age range, 18-85 years; median age, 58 years) with a median of two coronary artery disease risk factors.
Because of the concern for variation in coronary artery calcium quantification, every patient underwent two consecutive cardiac scans at one imaging session without being moved between scans to minimize any positional differences between scans. Up to 40 contiguous, 3-mm-thick transverse slices were obtained from approximately 2 cm inferior to the carina to the inferior margin of the heart. Each scan was obtained in a single breath-hold using a 100-msec exposure and the single-slice mode of imaging. ECG triggering was used to acquire each image during diastole corresponding to 80% of the R-R interval. Image reconstruction was performed with a 260-mm field of view using a 512 x 512 matrix and a sharp reconstruction filter to give a nominal pixel area of 0.26 mm2 and a voxel volume of 0.77 mm3.
A radiology technologist scored each of the scans. The technologist placed a region of interest around each calcific focus associated with an epicardial artery. The minimum criterion for a calcific lesion was two contiguous pixels with density equal to or greater than the accepted standard value of 130 H [6,7,11]. Originally described by Agatston et al. [6], the coronary artery calcium scoring algorithm that comes as part of the proprietary software package (version 12.2; Imatron, South San Francisco, CA) of the electron beam CT scanner assigns a numeric value between 1 and 4 on the basis of the peak pixel intensity of a calcific lesion. A lesion with a peak pixel intensity between 130 and 199 H is assigned a value of 1; a peak pixel intensity between 200 and 299 H is assigned a value of 2; a peak pixel intensity between 300 and 399 H is assigned a value of 3; and a lesion with a peak pixel intensity greater than or equal to 400 H is assigned a value of 4. The total area in square millimeters of each calcified lesion is determined by the software package and multiplied by the value derived from the peak pixel intensity to yield a calcium score for that lesion. The sum of all calcium scores for each of the major epicardial arteries is determined as part of the software analysis, as is the calcium score for the entire scan. In this study, a total coronary artery calcium score and the total number of vessels with calcifications (up to a maximum of four) were measured for each of the two scans. One of two radiologists evaluated the studies for technical quality and appropriateness of lesion identification and provided the final interpretation of the scans.
The possible clinical impact of the difference in calcium scores between scans was evaluated. Patients at our institution receive their clinical recommendations on the basis of their average calcium score and their cardiac risk factor profile. Standard cardiac risk factors (age, sex, serum lipid profile, body mass index, smoking habits, blood pressure, diabetes, family history, exercise frequency and duration, and dietary patterns) were assessed using a self-administered questionnaire given to the patient before scanning. Because different authors have reported different calcium score cutoff values to determine patient risk, we analyzed the effect of the cutoff value on the number of subjects in this asymptomatic population who would have been affected because of their interscan variation in calcium scores.
For all patients, the percentage of difference in calcium scores (as defined by the absolute value of the difference in calcium scores divided by the mean calcium score and expressed as a percentage) and the absolute value of the difference in calcium scores between scans were determined. The Wilcoxon's matched pairs test was used to evaluate the difference in scan results between the first and second scans. To determine the interscan correlation in calcium scores, the data were logtransformed (log10[calcium score + 1]) to reduce skewness. One-way analysis of variance was used to make comparisons of interscan variations in calcium scores between vessels with post hoc least significant difference analysis performed in those cases in which a statistically significant difference was determined by analysis of variance. All statistical calculations were performed using statistical software (SPSS, Chicago, IL).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In 311 patients, the calcium score from the first scan was greater than that of the second scan; and in 286 patients, the calcium score from the second scan was greater than that of the first. This difference was not statistically significant (p = 0.893). Among the 951 patients, 354 (37.2%) had identical scores from both scans. In the patients with identical scores, 314 (88.7%) had a score of 0. In the remaining 597 patients who did not have identical scores on their scans, 101 (16.9%) had a calcium score of 0 on one of the two scans.
We found excellent correlation (R2 = 0.97) between the log-transformed calcium scores of the two scans for all 951 patients, as shown in Figure 1. However, when we compared the percentage of difference between scans against the mean calcium score value, we noted considerable interscan variation despite the excellent correlation between scans. The percentage of difference was greatest among patients with low calcium scores (Fig. 2), but a few patients with scores greater than 500 had percentage of differences exceeding 40%.
|
|
Calcium score variations between scans for all 951 subjects are presented in Table 1 both as an average percentage of difference in scores and as the average absolute difference. One-way analysis of variance showed a significant difference in the percentage of difference in interscan variation among vessels for all patients (F score = 2.98, p = 0.03). Post hoc least significant difference analysis revealed significant differences between the left anterior descending and the circumflex arteries and between the left anterior descending and the right coronary arteries. However, when analysis of variance was limited to women, no significant interscan variation among vessels was seen. Analysis of variance of the interscan variation among vessels for men showed a significant difference (F score = 5.01, p = 0.002). Post hoc least significant difference analysis revealed significant differences between left main and right coronary arteries, left anterior descending and circumflex arteries, and left anterior descending and right coronary arteries. The greatest interscan variation was seen in the right coronary artery for men as well as for all patients. However, it was greatest in the left anterior descending artery for women.
|
Variation in the prevalence of calcium in the individual coronary arteries affects the calculated percentage of difference in interscan calcium scores. As shown in Table 2, the difference in interscan calcium scores was calculated only for those subjects in whom the specified vessel was noted to have calcium on at least one of their two scans. There is a large increase in average percentage of difference (as compared with the percentage of difference for all patients), as would be expected given the removal of all double-zero scores for each vessel. The left main artery then shows the greatest interscan variation in both men and women. One-way analysis of variance showed a significant difference in the interscan variation among vessels for all patients (F score = 16.9, p < 0.001). Post hoc least significant difference analysis revealed significant differences between the left anterior descending artery and the other three vessels. However, when analysis of variance was limited to women, no significant interscan variation among vessels was seen (F score = 1.9, p = 0.12). Analysis of variance of the interscan variation among men did show a significant difference (F score = 16.2, p < 0.001), and post hoc least significant difference analysis revealed significant differences between the left main artery and the other three coronary arteries.
|
Interscan variation in the vessel distribution of calcium was analyzed. Most subjects (66.7%) had no difference in the vessels identified with calcium on both scans, as shown in Table 3. This finding includes 314 subjects who did not have any calcium seen on either scan and 320 subjects in whom calcium was noted to be in the same vessels on both scans. For only two subjects did a discordance occur in all four vessels. One man had calcium noted in each of the four coronary arteries on the first scan and no calcium identified in any of these vessels on the second scan. This person had calcium scores of 9 and 0 on his first and second scans, respectively. The other subject had calcium noted in his left main, left anterior descending, and circumflex arteries on his first scan, with a combined calcium score of 5, but had calcium noted only in his right coronary artery on his second scan, with a total calcium score of 1.
|
The interscan variation in the overall number of coronary arteries that contained calcium was also evaluated (Table 3). Again, most patients had no difference between the two scans in the overall number of vessels noted to have calcium. Not unexpectedly, the Wilcoxon's test (p < 0.001) showed significantly less interscan difference in the overall number of vessels identified with calcium than the interscan variation in the individual vessels noted to have calcium.
When individual vessels were evaluated for the presence of calcium, the left anterior descending artery was the vessel most likely to have a calcific focus (seen in 47% of all subjects), followed closely by the right coronary artery (Table 4). Also, calcific foci were seen most consistently on both scans if located in the left anterior descending artery, contrasting with the right coronary artery, which had the lowest reproducibility of calcific lesions between scans. When the ratio of patients with a discrepancy in the presence of calcium in a particular vessel to the total number of patients with calcium seen in that vessel on either scan is evaluated, the left main artery shows the greatest percentage of discrepant cases.
|
To avoid confusion during the remainder of the analysis, the scan with the higher calcium score for each subject is referred to as SCAN1, and the scan with the lower calcium score as SCAN2. Among women, 164 had at least one scan with a calcium score of 0. Of these women, 140 (85.4%) had a calcium score of 0 on both scans, 21 women (12.8%) had a calcium score of 1 on SCAN1, and three (1.8%) had a calcium score greater than or equal to 2 on SCAN1, with the largest discrepancy being a calcium score of 4.
Greater discrepancy occurred in interscan calcium scores among 251 men with at least one calcium score of 0. Of these men, 174 men (69.3%) had a calcium score of 0 on both scans, 48 men (19.1%) had a calcium score of 1 on SCAN1, and 29 men (11.6%) had a calcium score greater than or equal to 2 (up to 9) on SCAN1.
The contribution of each vessel to the total calcium score for the 637 patients with at least one nonzero calcium score is shown in Table 5. The analysis was performed with SCAN1 data (the scan with the higher calcium score). The values represent the fraction of the total calcium score found for each vessel. In both sexes, the left main coronary artery makes the smallest and the left anterior descending artery the largest contribution to the total calcium score.
|
The most frequent site of coronary calcium in both older men and older women with calcium was the left anterior descending artery (Tables 6 and 7). A disproportionately greater increase was seen in the frequency of calcification of the left main and left anterior descending arteries than in the circumflex and right coronary arteries in the older age groups as compared with their younger counterparts for both sexes.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An additional interesting finding from our data is the change in the relative frequency of calcific lesions in the epicardial arteries associated with increasing age. We find that as patients get older, a greater likelihood exists that the left anterior descending and left main arteries will become involved with coronary calcification than the right coronary or circumflex artery. This observation holds true for both sexes. A similar finding was reported by Maher et al. [14], who found that age more closely related to increasing coronary calcium in the left main and left anterior descending arteries than in the right coronary or circumflex arteries for both men and women. This asynchronous pattern of coronary calcification may imply that atherosclerotic progression is not uniform among all the epicardial arteries. Alternatively, these findings may imply that atherosclerotic lesions may calcify at different rates over time. Hence, lesions of the right coronary and circumflex arteries may have a lower propensity to calcify with age than lesions involving the left anterior descending and left main arteries.
When interscan variation in calcium scores was limited to only those vessels in which calcium was identified in at least one of the two scans, the largest score variation occurred in the left main coronary artery. This finding may have been due in part to the difficulty in delineating the exact junction between the left main coronary artery and the left anterior descending and circumflex branches. A calcification near the junction of the left main bifurcation into the left anterior descending and circumflex arteries may be assigned to the left main artery, whereas on a second scan, the same lesion may be assigned to the left anterior descending or circumflex branch. Kaufmann et al. [7] reported the greatest intra- and interobserver variability in calcium scores for a single scan resulted from this same problem. Although this difficulty in assigning calcifications to a single artery should also increase the interscan variation for the left anterior descending and circumflex vessels, the more frequent involvement and greater degree of calcification of these vessels may obscure this effect.
Our study shows that interscan variation in calcium scores is an important concern when using electron beam CT in the quantification of coronary artery calcium. Although the existing software package determines both the calcium score and the lesion number, we chose not to evaluate interscan differences in lesion number. We would expect a greater variation in lesion number exists than in calcium scores because the current algorithm available on commercial electron beam CT scanners evaluates each image separately, so that a focus of calcification that spans several axial images is counted as a separate lesion on each image. This results in significant overestimation of lesion number. Less frequently, the number of calcific foci is underestimated if the user inadvertently chooses a region of interest that encompasses multiple closely approximated but distinct calcific foci. Although lesion number was not evaluated, the number of vessels with calcific foci was evaluated. Because the number of vessels involved with calcification has been shown to correlate with likelihood of significant stenosis [15], it seems appropriate that the number of vessels with calcifications continue to be reported. In our study, 293 (30.8%) subjects had an interscan difference of at least one coronary artery in which calcium was present on only one scan.
In this asymptomatic population, the absence of coronary artery calcification on one scan was rarely accompanied by a large amount of coronary artery calcium identified on a second scan. The largest difference in calcium scores was 4 in women and 9 in men among those with a calcium score of 0 on one of their two scans. Therefore, unless very low calcium score values are used to identify those at increased risk, it would seem prudent to reduce exposure to radiation by performing only a single scan on subjects who do not have calcium visualized on their first scan. Because image reconstruction of the first scan can be performed in a few minutes and no significant time is required to score a subject who has no calcific lesions, this practice should not significantly inconvenience the patient or technologist. Patients noted to have any calcific lesions on the first scan can undergo the second scan, with formal scoring of both scans performed after all imaging has been completed.
Neither the use of coronary artery calcium quantification nor the method of clinical risk stratification is universally accepted. The imaging and scoring methods used for the study, although admittedly not ideal, are those recommended by the vendor and most frequently reported by other authors [1, 3,4,5,6,7,8,9,10,11, 15,16,17]. Clinical risk stratification is even more controversial because outcomes data are relatively scarce. Arad et al. [1] found that using a calcium score of greater than 160 resulted in the maximum sensitivity (89%) and specificity (82%) for identifying asymptomatic patients at increased risk for cardiovascular events. In our study, 14 subjects (11 men and three women) would have been selected as being at increased risk for cardiovascular events only on the basis of their higher calcium score if a threshold calcium score of 160 were applied. These 14 patients represent 1.5% of the 951 patients, and the average difference in calcium score between the two scans in these 14 patients was 53.4.
It seems highly unlikely that a single value is applicable to all ages and both sexes given the pathogenesis and progression of atherosclerosis. Several other threshold values have been suggested. Rumberger et al. [16] found that in patients with no history of coronary artery disease, calcium score threshold values that maximized both sensitivity and specificity for detecting stenoses on angiography varied from 15 to 327, depending on the minimum stenosis severity that was detected.
In another study from the same group of researchers, Kaufmann et al. [18] examined a group of younger patients (23-59 years old) using both electron beam CT and coronary angiography. This group of researchers found that the median calcium score associated with mild disease (<50% angiographic stenosis) was 19, and the median calcium score associated with significant disease (>50% angiographic stenosis) was 327. These researchers used the calcium area rather than the calcium score to determine their sensitivity and specificity values.
In a multicenter study, Detrano et al. [19] showed that a calcium score threshold of 100 was a better predictor of subsequent cardiac events, with a sensitivity of 70% and specificity of 71%, than any other risk factor, including coronary angiographic findings. In our study, 26 (2.7%) subjects would have been selected as being at increased risk for subsequent cardiovascular events only on the basis of their higher score if we used a threshold calcium score of 100. The average difference in calcium scores between the two scans in these 21 men and five women was 44.3.
Fallavollita et al. [17] found that a calcium score threshold of 5 optimized sensitivity (59%) and specificity (87%) in predicting early (<50% diameter stenosis) atherosclerosis in 98 patients who had angiographic correlation. A total of 53 patients would have been categorized as being at increased risk only on the basis of their higher calcium scores if we had used a calcium score threshold of 5 in our study. The average difference in calcium scores between the higher and lower value scans among these 45 men and eight women was 5.4.
The population studied by Arad et al. [1] was most similar to our study population in that both groups of subjects were asymptomatic for coronary artery disease and were predominantly self-referred patients with multiple coronary risk factors. The results from Detrano et al. [19] and Fallavollita et al. [17] were obtained from patients who were undergoing coronary angiography. Hence, the pretest probability of coronary artery disease in those groups of patients would presumably be much higher than in an asymptomatic population. This may account for the lower calcium score cutoff values suggested by these latter two studies.
Clinical recommendations should not be made solely on the basis of calcium score values but should take into account individual risk factors and medical histories. Our retrospective study of an asymptomatic population appears similar in its prevalence of calcification to several other published series on asymptomatic subjects [20,21,22]. All showed strong association between the presence and extent of coronary calcium and patient age and sex. However, none of these studies had sufficient follow-up to allow the evaluation of the prognostic significance of the calcium score. As more data relating patient outcomes to calcium score become available, we expect that the risk stratification based on calcium score values will be considerably refined. In addition, new imaging and scoring algorithms have been proposed, and some appear quite promising [23, 24]. However, these studies represent small samples of selected populations and the scoring algorithms are not in general use. The purpose of this paper is not to suggest a particular method of risk stratification using calcium score values. Rather, we wished to show that interscan variation in calcium scores may have an impact on the clinical recommendations made to certain patients. The effect of interscan variations will depend on both the threshold values used to identify high-risk patients and the prevalence of coronary artery calcium in the study population.
In conclusion, interscan variation remains an important limitation of electron beam CT in the examination of asymptomatic patients. Patients without identifiable coronary calcifications on their first scan are not likely to benefit from an additional scan. However, in a minority of patients, a second scan may result in a sufficiently different calcium score to warrant a reappraisal of the clinical recommendation. The number of affected patients will be determined by both the calcium score threshold value selected by practitioners and by the prevalence of coronary artery calcium in the patient population. If the purpose of the screening protocol is to aggressively identify and treat patients who are at increased risk for cardiovascular events, the higher calcium score should be used for clinical risk stratification.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Rutten, I. Isgum, and M. Prokop Coronary Calcification: Effect of Small Variation of Scan Starting Position on Agatston, Volume, and Mass Scores Radiology, December 1, 2007; 246(1): 90 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. H. McCollough, S. Ulzheimer, S. S. Halliburton, K. Shanneik, R. D. White, and W. A. Kalender Coronary Artery Calcium: A Multi-institutional, Multimanufacturer International Standard for Quantification at Cardiac CT Radiology, May 1, 2007; 243(2): 527 - 538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Clouse, J. Chen, H. M. Krumholz, M. E. Clouse, J. Chen, and H. M. Krumholz Noninvasive Screening for Coronary Artery Disease With Computed Tomography Is Useful Circulation, January 3, 2006; 113(1): 125 - 146. [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. W. Sandstede, J. Stoffels, F. Wendel, C. Ritter, M. Beer, and D. Hahn Different Reconstruction Intervals for Exclusion of Coronary Artery Calcifications by Retrospectively Gated MDCT Am. J. Roentgenol., January 1, 2006; 186(1): 193 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Sevrukov, J. M. Bland, and G. T. Kondos Serial Electron Beam CT Measurements of Coronary Artery Calcium: Has Your Patient's Calcium Score Actually Changed? Am. J. Roentgenol., December 1, 2005; 185(6): 1546 - 1553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. J. Schoepf, C. R. Becker, B. M. Ohnesorge, and E. K. Yucel CT of Coronary Artery Disease Radiology, July 1, 2004; 232(1): 18 - 37. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Stanford, B. H. Thompson, T. L. Burns, S. D. Heery, and M. C. Burr Coronary Artery Calcium Quantification at Multi-Detector Row Helical CT versus Electron-Beam CT Radiology, February 1, 2004; 230(2): 397 - 402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Van Hoe, K. G. De Meerleer, P. Ph. Leyman, and P. K. Vanhoenacker Coronary Artery Calcium Scoring Using ECG-Gated Multidetector CT: Effect of Individually Optimized Image-Reconstruction Windows on Image Quality and Measurement Reproducibility Am. J. Roentgenol., October 1, 2003; 181(4): 1093 - 1100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Takahashi and K. T. Bae Quantification of Coronary Artery Calcium with Multi-Detector row CT: Assessing Interscan Variability with Different Tube Currents—Pilot Study Radiology, July 1, 2003; 228(1): 101 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Lu, N. Zhuang, S.-S. Mao, J. Child, S. Carson, H. Bakhsheshi, and M. J. Budoff EKG-triggered CT Data Acquisition to Reduce Variability in Coronary Arterial Calcium Score Radiology, September 1, 2002; 224(3): 838 - 844. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Valabhji, A. J. McColl, W. Richmond, M. Schachter, M. B. Rubens, and R. S. Elkeles Total Antioxidant Status and Coronary Artery Calcification in Type 1 Diabetes Diabetes Care, September 1, 2001; 24(9): 1608 - 1613. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. F. Bielak, P. F. Sheedy II, and P. A. Peyser Coronary Artery Calcification Measured at Electron-Beam CT: Agreement in Dual Scan Runs and Change over Time Radiology, January 1, 2001; 218(1): 224 - 229. [Abstract] [Full Text]
|
|
|
|
|
|
M. J. Lipton, L. M. Boxt, and Z. M. Hijazi Role of the Radiologist in Cardiac Diagnostic Imaging Am. J. Roentgenol., December 1, 2000; 175(6): 1495 - 1506. [Full Text] [PDF]
|
|
|
|
|
|
K. D. Hopper, D. C. Strollo, and D. T. Mauger Comparison of Electron-Beam and Ungated Helical CT in Detecting Coronary Arterial Calcification by Using a Working Heart Phantom and Artificial Coronary Arteries Radiology, February 1, 2002; 222(2): 474 - 482. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIO |
