HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Venema, H. W. Articles by den Heeten, G. J. Search for Related Content PubMed PubMed Citation Articles by Venema, H. W. Articles by den Heeten, G. J.

CT Angiography of the Circle of Willis and Intracranial Internal Carotid Arteries: Maximum Intensity Projection with Matched Mask Bone Elimination—Feasibility Study¹

¹ From the Departments of Radiology (H.W.V., F.J.H.H., G.J.d.H.) and Medical Physics (H.W.V.), Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands. From the 1999 RSNA scientific assembly. Received January 11, 2000; revision requested February 12; revision received July 11; accepted August 29. Address correspondence to H.W.V. (e-mail: h.w.venema@amc.uva.nl).

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
Results are reported of a feasibility study in which bone pixels are eliminated from computed tomographic (CT) angiographic images with a method that enables the construction of maximum intensity projection (MIP) images without interference by bone. The method proved to be successful in six patients. Two observers blinded to the bone elimination method judged the image quality of MIP images to be considerably higher than that of standard subtraction MIP images. This method is an effective means to remove bone from CT angiographic images with only a slight increase in radiation dose.

Index terms: Aneurysms, cerebral, 172.73, 175.73 • Cerebral angiography, technology • Computed tomography (CT), angiography, 172.12116, 175.12116 • Computed tomography (CT), maximum intensity projection • Images, processing

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
Computed tomographic (CT) angiography is an important imaging modality for the detection of cerebral aneurysms, arterial stenosis, and other vascular anomalies in the intracranial arteries. This technique is less invasive than digital subtraction angiography (DSA) because it requires only intravenous instead of intraarterial injection of contrast material. It can be performed with a substantially shorter examination time with less risk to the patient. CT angiography provides important three-dimensional (3D) information, which can be used as a navigational tool by neurosurgeons and interventional neuroradiologists.

Source images from CT angiography are often displayed as maximum intensity projection (MIP) reconstructions to obtain angiographic images (1,2). A drawback of the MIP technique is that, especially in the region of the skull base, extensive preprocessing is needed to remove the bone from the source images. Although some software tools are available to aid the operator in this task, preprocessing is time-consuming (2), and bone removal is often incomplete. This is especially the case in regions where arteries are contiguous with the bone, as in the petrous carotid region with its intricate mingling of bone and arteries (1). As a consequence, aneurysms at the skull base that arise from the intracavernous or supraclinoid carotid artery may be obscured by bone (3,4), and the assessment of stenosis in the petrous portion of the carotid artery is poor (5).

We developed a method, matched mask bone elimination (MMBE), to eliminate the bone pixels from CT angiographic source images, to create artifact-free MIP images of the intracranial arteries. Although MMBE has some resemblance to subtraction, it is fundamentally different and has substantial advantages. Our purpose was to evaluate the results of a feasibility study in which the effectiveness of this approach was evaluated.

	Materials and Methods

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
Data Acquisition
A feasibility study of the MMBE method was approved by our institutional review board, and informed consent was obtained from all patients. CT angiography was performed in six patients for the detection of cerebral aneurysms and other vascular anomalies. The six patients were adults (two men, four women; age range, 34–64 years; mean age, 50 years) who were consecutively admitted for CT angiography of the circle of Willis and the intracranial internal carotid arteries (Table 1). In each patient, two spiral CT scans of the same region were acquired by using a CT scanner (CT-Twin/Flash; Elscint, Haifa, Israel) with a double-detector array. A nonenhanced spiral CT scan was obtained, and then the CT angiographic examination was performed.

Nonenhanced CT scans in the first two patients were obtained with the same technique used in the CT angiographic examination. Bone elimination with the MMBE method appeared to be successful in these first two patients (see Results), confirming the validity of the approach. Because we wanted to keep the additional radiation dose as low as possible, we also investigated whether the method worked well with a low-dose technique for the nonenhanced scans. Therefore, in the next four patients, nonenhanced images were acquired with 67 mAs per section, that is, one-fourth of the milliampere-second value of the CT angiographic examination, with all other parameters being the same. In one of these patients, another CT angiographic examination had been performed 9 months earlier. These CT angiographic images were also evaluated with the MMBE method.

Reconstructions of both the nonenhanced and contrast material-enhanced scans were performed by using 360° linear interpolation with a 512 x 512 matrix, a pixel size of 0.29 mm, and a reconstruction interval of 0.5 mm, resulting in a stack of roughly 100 transverse images for each scan. All images were transferred to a personal computer equipped with a 350-MHz processor (Pentium Pro processor; Intel, Santa Clara, Calif) and processed with software we developed, as described later. The processing for the complete MMBE procedure was slightly more than 1 hour.

MMBE Method
The principle of MMBE is that bone pixels are identified in the nonenhanced data set and that the corresponding pixels in the CT angiographic data set are assigned an arbitrarily low value. MIP images free from overprojecting bone can then be obtained in the ordinary way.

The implementation was as follows. The two 3D data sets were matched. This procedure was necessary because sections obtained with identical table positions did not match perfectly due to slight movements of the patient between the scans (Fig 1, A–C). In the matching procedure, each image of the contrast-enhanced data set was registered with the 3D nonenhanced data set. A fully automatic 3D global rigid matching method was used on the basis of gray value correlation (6). With this method, the mean of the squared differences between the pixel values of each contrast-enhanced image and the corresponding pixel values in a cross-section of the 3D nonenhanced data set were minimized.

View larger version (133K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Comparison of regular subtraction, matched subtraction, and MMBE. A, Source image from a CT angiographic examination (265 mAs per section). B, Nonenhanced image obtained a few minutes earlier with an identical table position (67 mAs per section). C, Regular subtraction image (image in A minus image in B) reveals the amount of mismatch of both images. D, Matched nonenhanced image obtained by matching the image in A with the 3D data set of about 100 nonenhanced images. E, Matched subtraction image (image in A minus image in D). Note that although nonenhanced images in B and D appear virtually identical, the subtraction images are vastly different; image in C is of poor quality, and image in E is of good quality. This difference underlines the necessity of matching. F, Mask image obtained by applying a threshold to the image in D, in which all pixels in D with a CT value exceeding 150 HU were labeled (gray), followed by dilation with one pixel (white). G, MMBE image obtained by masking the CT angiographic image in A with the image in F. All pixels in A corresponding to gray or white pixels in F were set to a tissue-equivalent value (40 HU).

Next, the matched nonenhanced images were converted into mask images. A threshold of 150 HU was applied to identify the bone pixels. The mask was slightly widened by means of dilation (11) with 1 pixel (0.3 mm) to allow for partial-volume effects and slight amounts of mismatch (Fig 1, F). Finally, the pixels in the contrast images corresponding to the mask image pixels were set to an arbitrarily low value (40 HU) (Fig 1, G).

MIP images were obtained in the conventional way. MIP images were also made from sets of matched subtraction images (Fig 1, E) to evaluate the relative quality of bone elimination with MMBE and matched subtraction.

Evaluation of Images
In all six patients, MIP images were obtained from MMBE images in three directions (coronal, sagittal, and transverse planes), both for the complete field of view (15 x 15 cm) and for a smaller region of interest (ROI) (7.5 x 7.5 cm) centered on the internal carotid arteries. These images were assessed by two observers (F.J.H.H., G.J.d.H.) for the quality of the depiction of the vascular anatomy and pathologic findings in the region of the skull base for the presence of remnants of bone, and for artifacts. The observers also evaluated if there were differences in the quality of the MIP images of the first two patients (made with a regular dose mask) and of the other four patients (made with a low-dose mask).

The relative merits of MMBE and matched subtraction were evaluated in a study in which observers were blinded to the bone elimination method. Each of the 18 MIP images of the smaller ROI was printed on film next to a corresponding matched subtraction MIP image in a random order by using a laser imager (Matrix Compact L; Agfa-Gevaert, Mortsel, Belgium). These 18 pairs were judged by the same two observers according to three criteria: quality of delineation of the intracranial vessels, noise level, and presence of artifacts. The observers were asked to rank each pair of ROIs according to each criterion and to classify the differences, if present. A three-point scale was used as follows: 0 was no difference or only a slight difference, 1 was a difference, and 2 was a substantial difference.

As a consequence of this scoring system, the differences in quality between the MMBE images and the matched subtraction images were expressed on five-point scale that ranged from +2 (MMBE image substantially better) to -2 (matched subtraction image substantially better). For each criterion and for each observer, the mean score per patient was determined, and the significance of the difference of the mean of these mean scores from 0 was determined by using the Student t test.

	Results

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
Both observers independently judged the depiction of the vascular anatomy and pathologic findings in the region of the skull base the MIP images to be of high quality (Figs 2–4). Both observers were of the opinion that only clinically unimportant remnants of bone and artifacts were visible. They did not see any difference in the quality of the MIP images obtained with regular or with low-dose mask images.

View larger version (179K): [in this window] [in a new window] [Download PPT slide]   Figure 2. MIP images of an ROI of 256 x 256 pixels (7.5 x 7.5 cm) in the center of CT angiographic images obtained in patient 3, also shown in A and B, Without bone elimination. C and D, With matched subtraction. E and F, With MMBE. Images in A, C, and E are MIP images in the coronal plane; images B, D, and F are in the sagittal plane. Nonenhanced images used in subtraction and MMBE were obtained at 67 mAs per section, that is, one-fourth of the milliampere-second value at the CT angiographic examination. Both MMBE and subtraction reveal the vasculature in the skull base. MMBE images have a higher quality in the delineation of the arteries and show fewer artifacts due to remnants of bone. Comparison of upper parts of the corresponding MIP images shows that with MMBE (images in E and F) the image quality of the original MIP images (images in A and B) is retained; in the MIP images obtained with matched subtraction (images in C and D), it is degraded.

View larger version (152K): [in this window] [in a new window] [Download PPT slide]   Figure 3. MMBE MIP coronal images of the complete field of view (150-mm diameter) in patients 1-4. In A and B, regular-dose nonenhanced images were used; in C and D, low-dose nonenhanced images (obtained at one-fourth of the milliampere-second value) were used. In C, the same CT angiogram as in Figures 1 and 2 is shown. In the middle of the image in B, small artifacts are visible (arrow) due to the presence of a high-attenuating clip at the anterior communicating artery on the source images. This clip was removed by using the MMBE procedure.

View larger version (151K): [in this window] [in a new window] [Download PPT slide]   Figure 4. CT angiographic images obtained in February and October 1999 in patient 5. On the last occasion, a nonenhanced CT scan was also obtained. A-C, MIP images (October 1999) show the complete field of view (150-mm diameter) with the bone removed by using MMBE in the A, coronal; B, sagittal; and C, transverse planes. D-F, MIP images (February 1999) masked by using nonenhanced images from October 1999, in approximately the same projections as in A-C. A high-attenuating contrast material-filled balloon (arrow in D, E, and F), which was inserted in February 1999 to occlude the right vertebral artery, is visible. In October 1999 (images in A-C), the balloon has disappeared. The right vertebral artery is still occluded, however.

The MMBE technique proved to be a useful adjunct in the visualization of the vasculature of the four patients (three of which are shown in Figure 3) with a known aneurysm in the region of the skull base. Good-quality MIP images were also obtained at two CT angiographic examinations, 9 months apart, in one patient by using nonenhanced images obtained at the time of the last examination. The MIP images of this last examination are shown in Figure 4, A–C. The first CT angiographic examination was performed immediately after balloon occlusion of the vertebral artery (Fig 4, D–F). Bone removal was of definite value because the relevant vasculature was completely surrounded by the bone of the skull base. On a follow-up CT angiogram obtained 2 months after the occlusion (not shown), the balloon was still visible, but 7 months later it had disappeared (Fig 4, A–C). With all probability, it did deflate and became incorporated in the clot.

In this last example (with 9 months between CT angiographic examination and mask images), a slightly better match was obtained when the mandible was first removed from the images obtained at the CT angiographic examination. This was done by defining two seed points within the two parts of the mandible in one of the images and by applying a connectivity algorithm (1).

	Discussion

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 
The starting point for the introduction of a new bone-elimination procedure to improve MIP images was that the image quality of the CT angiographic images should not be degraded as a result of the application of the procedure. This precluded the use of an ordinary subtraction technique (12–17), as is evident from Figure 2. It is important to realize that in a subtraction image the noise of the two images is added.

As a consequence, the quality of the subtraction images, therefore, that of the MIP images, is similar to one acquired from source images obtained with half of the milliampere-second value or one-fifth of the milliampere-second value when the nonenhanced images are obtained with one-quarter the regular milliampere-second value (Fig 1). The smallest arteries that can be identified on the source images are generally not visible on a MIP image (5), and this problem will be aggravated if the image quality deteriorates because of the subtraction procedure.

With MMBE, the original quality of CT angiographic images is retained, and MIP images of the intracranial arteries are free of overprojecting bone. It made no difference whether the nonenhanced CT scans were acquired with a regular dose or a low dose. In all six patients, good visualization of the intracranial arteries on the MIP images was obtained. The superiority of MMBE over matched subtraction was confirmed by the results of the observer study (Table 2).

The bone-elimination procedure features two points: 3D image matching and bone removal according to a matched mask. Image matching is important, as the head of the patient is prone to movement, and minimal movements lead to serious artifacts on the processed images (Fig 1, A–C). Image matching is indispensable when CT scans obtained at different times are matched, as in the patient in our study for whom images from two CT angiographic examinations were available (Fig 4).

Besides bone, other high-attenuating structures, such as aneurysm clips and calcifications, are also removed from the MIP images in the masking procedure. In the case of clips, some beam-hardening artifacts remain visible in proximity to the clip (Fig 3, B). In our study, one small calcification was completely removed in one patient. Further investigations are needed to determine if MMBE can be a solution to problems associated with calcifications in the arterial wall, such as difficulties in the grading of arterial stenoses in the presence of more extensive calcifications (18).

For routine applications, the 1-hour computation time for the complete MMBE procedure for one CT angiographic image is long. When the algorithm is optimized for speed and when a faster processor is used, we estimate that the computation time can be reduced to 15 minutes or less.

In the literature (12–17), different subtraction techniques are described for the purpose of bone elimination. In all of these studies except one, image registration was used, either a global rigid matching method (14,15) or an elastic matching technique (12,13,16). This last method allows for movements of the structures within the volume between the scans. MMBE can also be used with an elastic matching technique. Because of the rigidity of the skull, however, a global rigid matching method produced excellent results. The only movement of any importance within the volume depicted in the CT angiographic study that we noticed was a slight movement of the mandible relative to the skull between scans; this occurred in only the patient with two examinations performed 9 months apart. This scan was centered at the skull base and posterior fossa and included a substantial part of the temporomandibular joint. In the other patients, the scans were centered at a slightly higher level, and the motion of the mandible was not a problem. At present, we instruct our patients to keep their jaws closed during both nonenhanced and contrast-enhanced examinations. When images that contain bone structures that can move in relation to each other (eg, in the region of the upper cervical spine or craniocervical junction) are matched, the MMBE method can easily be extended by using matching for each bone structure separately (19).

To our knowledge, all techniques described in the literature make use of subtraction for bone elimination, with the drawbacks described previously. An advantage of the MMBE technique is that low-dose nonenhanced images can be used, as the noise on the nonenhanced images is not added to that of the CT angiographic images. Moreover, the nonenhanced images can be stored for future use in case repeat CT angiographic examinations are indicated. We conclude that MMBE is a useful adjunct for the display of CT angiograms; it can be used to remove the bone effectively while retaining the quality of the source images with only a modest increase in radiation dose.

	ACKNOWLEDGMENTS

 
We thank Charles B. L. M. Majoie, MD, PhD, and Edwin J. R. van Beek, MD, PhD, for critically reading the manuscript and making helpful comments.

	FOOTNOTES

 
Abbreviations: DSA = digital subtraction angiography, MIP = maximum intensity projection, MMBE = matched mask bone elimination, ROI = region of interest, 3D = three-dimensional

Author contributions: Guarantor of integrity of entire study, H.W.V.; study concepts, H.W.V., F.J.H.H.; study design, H.W.V., F.J.H.H., G.J.d.H.; definition of intellectual content, H.W.V., F.J.H.H., G.J.d.H.; literature research, H.W.V., F.J.H.H.; clinical studies, F.J.H.H.; data acquisition, H.W.V., F.J.H.H.; data analysis, H.W.V., F.J.H.H., G.J.d.H.; manuscript preparation and editing, H.W.V.; manuscript review and final version approval, H.W.V., F.J.H.H., G.J.d.H.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and Methods Results Discussion REFERENCES

 

1. Napel S, Marks MP, Rubin GD, et al. CT angiography with spiral CT and maximum intensity projection. Radiology 1992; 185:607-610.[Abstract/Free Full Text]
2. Velthuis BK, van Leeuwen MS, Witkamp TD, Boomstra S, Ramos LMP, Rinkel GJE. CT angiography: source images and postprocessing techniques in the detection of cerebral aneurysms. AJR Am J Roentgenol 1997; 169:1411-1417.[Free Full Text]
3. Schwartz RB, Tice HM, Hooten SM, Hsu L, Stieg PE. Evaluation of cerebral aneurysms with helical CT: correlation with conventional angiography and MR angiography. Radiology 1994; 192:717-722.[Abstract/Free Full Text]
4. Korogi Y, Takahashi M, Katada K, et al. Intracranial aneurysms: detection with three-dimensional CT angiography with volume rendering—comparison with conventional angiographic and surgical findings. Radiology 1999; 211:497-506.[Abstract/Free Full Text]
5. Skutta B, Fürst G, Eilers J, Ferbert A, Kuhn FP. Intracranial stenoocclusive disease: double-detector helical CT angiography versus digital subtraction angiography. AJNR Am J Neuroradiol 1999; 20:791-799.[Abstract/Free Full Text]
6. Van den Elsen PA, Pol EJD, Viergever MA. Medical image matching: a review with classification. IEEE Eng Med Biol Mag 1993; 12:26-39.
7. Nelder JA, Mead R. A simplex method for function minimization. Comput J 1965; 7:308-313.
8. Press WH, Flannery BP, Teukolsky SA, Vetterling WT. Numerical recipes Cambridge, Mass: Cambridge University Press, 1986; 289-293.
9. van Herk M, Kooy HM. Automatic three-dimensional correlation of CT-CT, CT-MRI, and CT-SPECT using chamfer matching. Med Phys 1994; 21:1163-1178.[Medline]
10. Venema HW, Phoa SSKS, Mirck PGB, Hulsmans FJH, Majoie CBLM, Verbeeten B, Jr. Petrosal bone: coronal reconstructions from axial spiral CT data obtained with 0.5-mm collimation can replace direct coronal sequential CT scans. Radiology 1999; 213:375-382.[Abstract/Free Full Text]
11. Gonzalez RC, Woods RE. Digital image processing Reading, Mass: Addison-Wesley, 1992; 518-524.
12. Beier J, Richter CS, Fleck E, Felix R. Image registration for subtraction angiography in spiral CT. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1994; 162:58-64.
13. Beier J, Oellinger H, Richter CS, Fleck E, Felix R. Registered image subtraction for CT-, MR- and coronary angiography. Eur Radiol 1997; 7:82-89.[Medline]
14. Görzer H, Heimberger K, Schindler E. Spiral CT angiography with digital subtraction of extra- and intracranial vessels. J Comput Assist Tomogr 1994; 18:839-841.[Medline]
15. Englmeier KH, Fink U, Haubner M, et al. A method of computer-assisted, 3-dimensional subtraction angiography using spiral roentgen computerized tomography. Radiologe 1996; 36:360-364.[Medline]
16. Luboldt W, Stehling MK, Pearlman JD, Seemann M, Raptopoulos V. CT subtraction angiography (CTSA). Results with an automated "elastic" subtraction algorithm. Radiologe 1997; 31:89-93.
17. Imakita S, Onisji Y, Hashimoto T, et al. Subtraction CT-angiography with controlled orbit helical scanning for detection of intracranial aneurysms. AJNR Am J Neuroradiol 1998; 19:291-295.[Abstract]
18. Bluemke DA, Chambers TP. Spiral CT angiography: an alternative to conventional angiography (editorial). Radiology 1995; 195:317-319.[Free Full Text]
19. Snel JG, Venema HW, Grimbergen CA, Moojen TM, Ritt MJ, Den Heeten GJ. Quantitative in vivo analysis of the kinematics of carpal bones using a deformable surface model and a 3D matching technique. Med Phys 2000; 27:2037-2047.[Medline]

This article has been cited by other articles:

				Y. Watanabe, N. Kashiwagi, N. Yamada, M. Higashi, T. Fukuda, S. Morikawa, Y. Onishi, K. Iihara, S. Miyamoto, and H. Naito Subtraction 3D CT Angiography with the Orbital Synchronized Helical Scan Technique for the Evaluation of Postoperative Cerebral Aneurysms Treated with Cobalt-Alloy Clips AJNR Am. J. Neuroradiol., June 1, 2008; 29(6): 1071 - 1075. [Abstract] [Full Text] [PDF]

				R. Manniesing, M. A. Viergever, A. van der Lugt, and W. J. Niessen Cerebral Arteries: Fully Automated Segmentation from CT Angiography--A Feasibility Study Radiology, June 1, 2008; 247(3): 841 - 846. [Abstract] [Full Text] [PDF]

				M. Romijn, H.A.F. G. van Andel, M.A. van Walderveen, M.E. Sprengers, J.C. van Rijn, W.J. van Rooij, H.W. Venema, C.A. Grimbergen, G.J. den Heeten, and C.B. Majoie Diagnostic Accuracy of CT Angiography with Matched Mask Bone Elimination for Detection of Intracranial Aneurysms: Comparison with Digital Subtraction Angiography and 3D Rotational Angiography AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 134 - 139. [Abstract] [Full Text] [PDF]

				M.M. Lell, H. Ditt, C. Panknin, J.W. Sayre, S.G. Ruehm, E. Klotz, B.F. Tomandl, and J.P. Villablanca Bone-Subtraction CT Angiography: Evaluation of Two Different Fully Automated Image-Registration Procedures for Interscan Motion Compensation AJNR Am. J. Neuroradiol., August 1, 2007; 28(7): 1362 - 1368. [Abstract] [Full Text] [PDF]

				M. M. Lell, K. Anders, M. Uder, E. Klotz, H. Ditt, F. Vega-Higuera, T. Boskamp, W. A. Bautz, and B. F. Tomandl New Techniques in CT Angiography RadioGraphics, October 1, 2006; 26(suppl_1): S45 - S62. [Abstract] [Full Text] [PDF]

				B.F. Tomandl, T. Hammen, E. Klotz, H. Ditt, B. Stemper, and M. Lell Bone-Subtraction CT Angiography for the Evaluation of Intracranial Aneurysms AJNR Am. J. Neuroradiol., January 1, 2006; 27(1): 55 - 59. [Abstract] [Full Text] [PDF]

				I. van der Schaaf, M. van Leeuwen, A. Vlassenbroek, and B. Velthuis Minimizing Clip Artifacts in Multi CT Angiography of Clipped Patients AJNR Am. J. Neuroradiol., January 1, 2006; 27(1): 60 - 66. [Abstract] [Full Text] [PDF]

				C. B. L. M. Majoie, M. van Straten, H. W. Venema, and G. J. den Heeten Multisection CT Venography of the Dural Sinuses and Cerebral Veins by Using Matched Mask Bone Elimination AJNR Am. J. Neuroradiol., May 1, 2004; 25(5): 787 - 791. [Abstract] [Full Text] [PDF]

				H. W. Venema, G. J. d. Heeten, and V. K. Jayakrishnan Subtraction Helical CT Angiography of Intra- and Extracranial Vessels: Technical Considerations and Preliminary Experience--Rediscovery of Matched Mask Bone Elimination? AJNR Am. J. Neuroradiol., August 1, 2003; 24(7): 1491 - 1492. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Venema, H. W. Articles by den Heeten, G. J. Search for Related Content PubMed PubMed Citation Articles by Venema, H. W. Articles by den Heeten, G. J.