|Year : 2016 | Volume
| Issue : 2 | Page : 31-39
A comparison of cone-beam computed tomography image quality obtained in phantoms with different fields of view, voxel size, and angular rotation for iCAT NG
Dorothy A Sonya1, Jonathan Davies2, Nancy Lee Ford3
1 Department of Dental Radiological Imaging, King's College London Dental Institute, London, UK; Department of Oral Biological and Medical Sciences, University of British Columbia, Vancouver, BC, Canada
2 Department of Dental Radiological Imaging, King's College London Dental Institute, London, UK
3 Department of Oral Biological and Medical Sciences, University of British Columbia, Vancouver, BC, Canada
|Date of Web Publication||10-Jun-2016|
Nancy Lee Ford
Department of Oral Biological and Medical Sciences, University of British Columbia, 2405 Wesbrook Mall, Room B210, Vancouver, BC, V6T1Z3, Canada
Source of Support: None, Conflict of Interest: None
Objectives: To characterize cone-beam computed tomography (CBCT) image quality (IQ) and dose for different fields of view (FOVs), voxel size, and angular rotation. Materials and Methods: IQ parameters, including image noise, homogeneity, geometric distortion, artifacts, contrast resolution and spatial resolution, and radiation dose, were measured for different FOV, voxel size, and angular rotation for an iCAT NG CBCT machine. Results: Noise increased with smaller voxel sizes as measured in the homogeneity layers. The 360° gantry rotation leads to improvements in contrast-to-noise ratio and spatial resolution and a decrease in artifacts compare to 180° gantry rotation with the same voxel size and FOV. Dose reduction was not always observed with smaller FOVs if smaller voxel sizes and longer scan times are used. Some of the test objects included in the phantom are not useful for dental CBCT machines, such as the range of resolutions tested by the bar pattern insert and the suitability of the materials used for the contrast assessment layer. Conclusions: A reduction in the patient dose can be achieved by reducing the angular rotation to 180°, increasing the acquired voxel size or decreasing the FOV height. However, using the reduced rotation angle also leads to increased artifacts around metallic objects. Changing the voxel size did not necessarily lead to improved spatial resolution or reduced dose, as some of the voxel sizes on this machine have identical imaging parameters.
Keywords: Cone-beam computed tomography, image quality, radiation dosimetry, radiographic phantom, radiography
|How to cite this article:|
Sonya DA, Davies J, Ford NL. A comparison of cone-beam computed tomography image quality obtained in phantoms with different fields of view, voxel size, and angular rotation for iCAT NG. J Oral Maxillofac Radiol 2016;4:31-9
|How to cite this URL:|
Sonya DA, Davies J, Ford NL. A comparison of cone-beam computed tomography image quality obtained in phantoms with different fields of view, voxel size, and angular rotation for iCAT NG. J Oral Maxillofac Radiol [serial online] 2016 [cited 2020 Jul 2];4:31-9. Available from: http://www.joomr.org/text.asp?2016/4/2/31/183821
| Introduction|| |
Over the past 10-15 years, the dental profession has rapidly incorporated cone-beam computed tomography (CBCT) technology into dental practices, universities, and hospitals with well over 3000 CBCT machines installed in the United States alone.  Manufacturers have responded with an increase in the number of models available to the dental profession with a wide range of imaging parameters, including field of view (FOV) size, beam quality (beam energy and filtration), radiation flux (anode current and exposure times), resolution, and rotation arc (half or full). Horner et al. have recently suggested that quality assurance programs require "suspension levels" to identify when machines are not compliant with guidelines or manufacturer specifications.  To develop guidelines and suspension levels, a wide-scale audit of image quality (IQ) and dosimetry for a range of CBCT machines must be reported.
IQ and radiation dose have been reported in the literature using various nonstandardized "subjects" such as skulls, and patient data or anthropomorphic phantoms. ,, This variety in CBCT imaging parameters and subjects used has made it difficult to accurately and easily compare the different CBCT machines. For a large scale audit, standardized phantoms must be utilized, such as those developed by Pauwels et al. In their study, seven different models were tested using the SEDENTEXCT phantoms, whereas Bamba et al. compared the measurements obtained in three different CBCT machines. Choi et al. used the same phantoms to compare objective measures of IQ with subjective metrics of diagnostic quality from an observer study. In these published studies, only a single FOV and at most two exposure settings per CBCT machine were reported. To fully characterize a CBCT system, and create a baseline range of IQ metrics suitable for inclusion in a large-scale audit, all available imaging FOV should be reported.
In this investigation, we measured five basic IQ parameters and the absorbed dose to compare different imaging FOV sizes, reconstructed voxel size, and angular rotation (180° or 360°) on a single CBCT unit. IQ and dosimetry were measured using standardized IQ and dose index (DI) phantoms developed by the SEDENTEXCT Consortium. 
| Materials and Methods|| |
Phantoms and dosimetry equipment
The IQ and DI phantoms (SedentexCT IQ, Leeds Test Objects Ltd., Boroughbridge, UK) are polymethyl methacrylate (PMMA) cylinders, 160 mm in diameter, and 162 mm in height, which approximate the radiographic properties of an adult head. Both phantoms can be mounted on a tripod for positioning in the FOV [Figure 1].
|Figure 8: Modulation transfer function curves for the 16 cm × 6 cm field of view for each voxel size obtained with (a) 180° scans and (b) 360° scans. For the 360° scans, the curves are shifted to higher modulation transfer function values at the same spatial frequency|
Click here to view
The bottom 22 mm of the IQ phantom contains solid PMMA for uniformity measurements and a plate containing a grid pattern of holes drilled in the PMMA spaced 10 mm apart to assess geometric distortion. The top 140 mm has six vertical columns around the periphery and one in the center containing inserts for measuring IQ, including image noise, artifacts, contrast resolution, and the limiting spatial resolution. Each set of test objects is located in a different imaging plane stacked vertically within the phantom. CBCT images of the different test objects are shown in [Figure 2]. Image noise is measured in PMMA plugs that fill all of the extra spaces within the columns above and below the test objects or in the bottom layer of PMMA; for this study, we used the columns to ensure consistent placement of the measurement locations between settings. The artifact insert contained three closely spaced titanium rods in two different columns, positioned in perpendicular rows. Contrast resolution is assessed using different materials (air, aluminum, polytetrafluoroethylene [PTFE], Delrin, and low-density polyethylene) suspended in PMMA. Each material has five rods of different diameters (1.0-5.0 mm) arranged in a circle in a single column, with all of the materials positioned in the same axial plane of the phantom. Spatial resolution is measured with bar patterns in the axial plane and z-axis, with the pattern spacing reported as 1.0, 1.7, 2.0, 2.5, 2.8, 4.0, and 5.0 lp/mm by the phantom manufacturer. Spatial resolution can also be calculated with the modulation transfer function (MTF) using either a slanted edge made of PTFE or a 0.25 mm thick stainless steel wire suspended in air.
The DI Phantom is made of solid PMMA slabs of differing thicknesses and has inserts to accommodate different sizes of ionization chambers, thermoluminescent detectors, or radiochromic film. The phantom has holes drilled through one set of slabs for the ionization chamber and PMMA plugs to fill the unused holes. To measure the radiation dose, a thimble ionization chamber (10X6-0.6-CT, Radcal Corporation, Monrovia, USA) was positioned in the central hole and connected to an Accudose digitizer (Accu-Dose, Radcal Corporation, Monrovia CA, USA). Exposure values were obtained using the high-sensitivity setting for improved accuracy and corrected for temperature and pressure and converted to absorbed dose to the PMMA phantom. The ion chamber was factory calibrated to be within ±5% at the energies used in CBCT.
|Figure 7: Images of the wire insert for the 16 cm × 6 cm field of view obtained at 0.3 mm voxel size. (a) In the 180° scan (4.8 s exposure time), the wire has a crescent shape, whereas in (b) the 360° scan (8.9 s exposure time), the wire appears more circular. The insert diameter is 3.5 cm, and the wire diameter is 0.25 mm|
Click here to view
Cone-beam computed tomography machine and imaging parameters
The CBCT machine used in this study was an iCAT Next Generation (Imaging Sciences International, Hatfield Pennsylvania, USA). Acquisition protocols were correlated to clinical situations and classified according to craniofacial FOVs and dentoalveolar FOVs. Only the FOVs that were most commonly used were tested in this study. The imaging parameters are summarized in [Table 1] and are the manufacturers default settings that are selected from a dropdown list on the operator console.
|Table 7: Absorbed dose values measured in the polymethyl methacrylate dose index phantom for the different scanning protocols for the iCAT next generation|
Click here to view
Each phantom was positioned in the scanner using the laser guides to align the phantom within the FOV. For the craniofacial images, the entire IQ phantom was imaged in a single acquisition, whereas the dentoalveolar images required multiple acquisitions to capture the entire IQ Phantom. For the iCAT NG machine, all FOV are centered on the phantom with a constant diameter of 16 cm and varying scan heights. To obtain the dentoalveolar images, the phantom was moved vertically between images using a crank on the tripod to align the different layers with the FOV.
Image quality phantom data analysis
The DICOM images were imported into ImageJ (1.45a, National Institutes of Health, Bethesda, USA) for analysis. For subjective measurement, a single observer viewed the images using the same monitor under constant viewing conditions and was permitted to enlarge the image if desired. Image uniformity and noise measurements were obtained by drawing circular regions of interest (ROIs) nearly covering the entire 3.5 cm diameter of each insert column and ensuring that no air was included at the periphery of the column, and calculating the mean gray value (uniformity) and the standard deviation (noise). Quantitative assessment of the uniformity was calculated using Equation 1. The uniformity was visually assessed by plotting a line across the uniform PMMA layer comprising the entire diameter of the phantom.
Image noise is the average of the standard deviations measured in PMMA of the seven insert columns. Geometric distortions were assessed in the voids region of the phantom by plotting 2 orthogonal lines and calculating the intervoid spacing using a custom script in MATLAB (Mathworks, Natick MA, USA). The script was programmed to identify the local minima along the line, calculate the distance between each pair of minima, and return the average spacing and standard deviation. In the artifact analysis layer, ROIs were drawn in all columns and the standard deviations compared to identify where the artifacts were most prominent. In this layer, two of the columns contained titanium rods embedded in the PMMA, and the remaining columns were filled with PMMA. The ROI drawn on the titanium rods included both materials within the 3.5 cm diameter. Contrast resolution was measured by taking the mean gray values in ROIs covering different materials to calculate the contrast-to-noise ratio (CNR) [Equation 2] and by subjectively identifying which materials were visible for each rod diameter.
Finally, the spatial resolution was obtained subjectively by visually inspecting the bar patterns to determine which spacing of lines were visualized and objectively by calculating the MTF. The MTF was calculated using MATLAB (Mathworks, Natick, MA, USA) by drawing a 30 × 30 pixel ROI on the wire insert (containing a 0.25 mm thick stainless steel wire suspended in air), and integrating to yield the point spread function (PSF). The PSF for 10 adjacent slices was obtained and averaged to reduce the noise, and the MTF was obtained using a fast Fourier transform.
The Radcal thimble chamber was placed into the centered column of the DI Phantom and the FOV was centered on this chamber using a scout scan. Three scans were taken using each of the same protocols used for the IQ phantom, and the measured exposure values were averaged and converted into absorbed dose to the PMMA. Dose-area product (DAP) values reported by the iCAT NG machine for each of the protocols were also recorded.
| Results|| |
Image quality phantom results
The variation in the mean values measured in the peripheral columns compared with the central column is shown in [Table 2], along with the uniformity metric. Although there was a tendency for lower values toward the middle of the phantom, the uniformity was excellent for all settings with most measurements exceeding 95% (ranging from 90% to 99%). The uniformity improved with the 360° acquisition (longer scans). A cupping artifact was noted in the middle of the individual plots for each scan time regardless of the voxel size, with 180° scans showing greater cupping [Figure 3].
|Figure 6: Examples of different line pair images for the 16 cm × 6 cm field of view in the axial plane: (a) 0.4 mm voxel, 8.9 s, (b) 0.3 mm voxel, 8.9 s, (c) 0.25 mm voxel, 14.7 s and (d) 0.2 mm voxel, 14.7 s. In all cases, the limiting resolution is 1 lp/mm, corresponding to the first grouping, although the spaces are easier to visualize for the smaller voxel sizes|
Click here to view
The image noise, shown in [Table 3], increased mostly for smaller voxel sizes and shorter scan times (180° acquisitions). To measure geometric distortion, two perpendicular lines were drawn across the center of the voids [Figure 4]. The data demonstrated excellent accuracy in the distance between the voids [Table 3] and for measurements of the diameter of the phantom. Linear measurements were very consistent for this layer with measurement accuracy within 1% of the manufacturer's specifications for this phantom.
|Figure 5: Artifact induction layer for the 16 cm × 6 cm field of view. The 0.4 mm voxel size with (a) 4.7 s exposure, 180° rotation and (b) 8.9 s exposure, 360° rotation and the 0.2 mm voxel size with (c) 14.7 s exposure, 180° rotation and (d) 26.9 s exposure, 360° rotation. Note the increased streak artifacts in the half-scan images (180° rotation)|
Click here to view
|Table 6: Spatial resolution metrics: Full-width at half-maximum of the 0.25 mm diameter wire measured in the axial plane, the 10% modulation transfer function in the axial plane and the corresponding spatial resolution of the imaging system|
Click here to view
|Table 5: Contrast to noise ratios for each of the five materials in the contrast resolution insert|
Click here to view
Artifacts appeared as streaks radiating from the titanium rods in the phantom. The PMMA insert located between the two artifacts induction inserts, the central column, was more affected than the four other peripheral columns [Figure 5], indicating that the artifacts were localized along the X-ray paths that crossed both metallic inserts, leading to photon starvation in the central column. The streaking was more pronounced for the half-scan 180° images corresponding to shorter scan times. The half-scan 180° images exhibited a different rod shape, with bright spoke-like structures streaking away from the rod making a more oblong shape, in contrast to the full 360° images, where the rods appeared as continuous round circles.
|Figure 4: Line profiles were used to assess the accuracy of linear measurements and the spacing between adjacent voids. (a) Polymethyl methacrylate slab containing the voids with 2 perpendicular lines drawn. (b) Plot across the voids|
Click here to view
To assess the contrast resolution, we used visual analysis to determine the smallest rod diameter visible in each of the five contrast resolution inserts for the different materials. Results tabulated in [Table 4] shows that Delrin was the most difficult to discern at smaller diameters, as expected. Shorter scan times sometimes led to reduced conspicuity of the smaller rods.
|Table 4: Data for visual analysis of the smallest rod size (in mm) that was clearly visible for the five materials in the contrast resolution insert for each of the scanning protocols. The phantom specifications for the rods range from 1 to 5 mm in diameter|
Click here to view
Pixel intensity values were averaged for each material used to calculate the CNRs [Table 5]. Pixel intensity values of the materials were different for each FOV setting indicating that the images are scaled differently. The relative CNRs were fairly consistent for all the scanning protocols, with similar trends between the materials with the lowest CNR observed for the Delrin, which is consistent with the results of the contrast resolution test in [Table 4].
Comparisons were only made across the protocols with the same voxel size since the image noise changes as a function of voxel size. Images obtained with shorter exposure times, while holding the FOV size and voxel size constant, yielded reduced CNR due to the increase in image noise. The CNR seemed to be improved for the smaller FOV sizes when comparing the same exposure time and voxel size.
|Table 3: Measured values for image noise and the mean intervoid spacing as a measure of geometric distortion. The intervoid spacing specified by the manufacturer is 10.0 mm|
Click here to view
Visual analysis of the bar patterns determined only the 1.0 lp/mm grouping could be clearly seen in the axial plane [Figure 6] for all images, with slight improvement for smaller voxels. None of the other groupings measuring 1.7, 2.0, 2.5, 2.8, 4.0, and 5.0 lp/mm were discernable as separated line pairs. Neither the voxel size nor the FOV size had an impact on the limiting axial resolution observed with the bar pattern method. However, the z-axis resolution was improved for smaller FOV sizes; for the 17 cm × 23 cm FOV, none of the resolution bar pattern groupings could be clearly resolved, but the 1.0 lp/mm was observed for all images with 16 cm ×13 cm and 16 cm × 6 cm FOVs. [Table 6] summarizes the axial spatial resolution metrics obtained from the full-width at half-maximum (FWHM) of the PSF and the 10% MTF. The 180° scans exhibit reconstruction artifacts that cause the wire to have an irregular shape compared to the circular appearance in the 360° scans [Figure 7], which resulted in larger FWHM and reduced MTF, as demonstrated by the MTF curves for the 16 cm × 6 cm FOV shown in [Figure 8].
|Figure 3: Plot across the uniform polymethyl methacrylate section of the image quality phantom showing the pixel-to-pixel variation and the cupping artifact, where the gray scale values in the center of the phantom are lower than those around the outer edge|
Click here to view
|Figure 2: Cone-beam computed tomography images of the image quality phantom depicting the different image quality inserts (a) homogeneous polymethyl methacrylate layer, (b) geometric distortion, (c) noise, (d) artifact induction, (e) contrast resolution, and (f) spatial resolution. The diameter of the phantom is 160 mm|
Click here to view
|Figure 1: SEDENTEXCT phantoms: (a) Side view of the SEDENTEXCT image quality phantom showing the different horizontal layers and peripheral columns containing test objects, and (b) SEDENTEXCT dose index phantom with the ion chamber inserted into the central hole|
Click here to view
|Table 2: Uniformity: Average gray values for polymethyl methacrylate in the central column of the phantom and the peripheral columns with the maximum and minimum mean gray values, with uniformity expressed as a percentage|
Click here to view
Dose index phantom results
The data for radiation dose are shown in [Table 7]. In general, both the DAP and dose values increased for smaller voxel sizes and longer scan times but not necessarily for larger FOV sizes. The 17 cm × 23 cm FOV had lower dose values than the other FOVs with identical mAs and voxel size; this is due to the 17 cm × 23 cm mode using 180°, whereas the other FOVs use a full 360° rotation for the same scan time, and a different acquisition geometry, where the detector is rotated 90° and an off-axis scanning technique is employed, reducing the radiation field size at each projection angle. Similarities between DAP and dose values for scans acquired at 0.3 mm and 0.4 mm voxel size as well as scans acquired at 0.2 mm and 0.25 mm voxel size were observed, indicating these pairs of scans are actually acquired with the same spatial resolution and that voxel size refers to postprocessing of the images for viewing. [Table 7] also shows an increase in dose for the 16 × 6 and 16 × 13 FOV for the 0.2 and 0.25 mm voxels compared with the 17 cm × 23 cm FOV measurements.
|Table 1: Scanning protocols for iCAT next generation cone beam computed tomography machine|
Click here to view
| Discussion|| |
In this study, three parameters were varied for an iCAT NG CBCT machine: Voxel size, FOV, and angular sampling of projections (180° or full 360° rotation). Overall, the images obtained exhibited good IQ, with measured values comparing well with other models of CBCT machines using the same phantom,  and repeated measurements yielded reproducible results. Images were uniform and the artifacts due to added metal were localized. The measured resolution was within the suggested guideline of 1 lp/mm  in the axial plane for all settings, and in the z-axis for the 16 cm × 13 cm and 16 cm × 6 cm FOV sizes. Linear measurements on the acquired images were accurate, suggesting patient images would correlate well with actual patient measurements, ,,, an important consideration in orthodontic treatment planning.
Several voxel sizes (0.4 and 0.3 mm and 0.25 and 0.2 mm) were obtained with the same mAs and DAP values but reconstructed with different voxel sizes; our measurements of dose and spatial resolution both indicate no differences in these images, with only small differences in the image noise. Using the MTF, the 0.2 and 0.25 mm settings exhibited better resolution than 0.3 and 0.4 mm acquisitions. For high contrast tasks, like discriminating between bone and soft tissue or between tooth and air, the smaller voxels may provide increased confidence for visualizing small details. The dose did increase for smaller voxel sizes, although the dose for the 0.3 and 0.4 mm spacing was identical, as was the dose for the 0.2 and 0.25 mm spacing. For this CBCT machine, it is not clear that changing the voxel size between 0.2 and 0.25 mm or between 0.3 and 0.4 mm is effective, as the spatial resolution and dose remain unchanged between these settings, with only small differences measured for the noise. Therefore, for this CBCT machine, selecting a larger voxel size would not necessarily reduce patient dose as is often suggested. ,
FOV selection should be based entirely on diagnostic task as there is little change in the IQ between FOV sizes although the z-axis resolution did improve for the smaller FOVs, which may be related to reduced X-ray scatter. For similar imaging settings, the dose was also reduced in the small FOV although FOV size by itself does not determine the radiation dose a patient might receive. ,,,,, For this machine, the kVp and mA are fixed for all imaging protocols and exposure times are the same for different FOVs, which is not true for many other CBCT models.
Angular spacing (180° or 360°) gives different scan times for each FOV and voxel size. Using the 360° acquisition resulted in longer scan times, corresponding to a higher radiation dose. However, this acquisition mode also had improved contrast, particularly for discriminating between tissues with similar attenuation properties, reduced artifacts near metal, and an improvement in the spatial resolution. For patients with metallic implants nearby the ROI or where soft tissue contrast is essential, using the 360° acquisition mode, associated with a longer scan time, would improve the IQ and may justify the additional radiation dose to the patient. The 360° mode may also be needed to accurately visualize the shape of small objects near the resolution limit, as [Figure 7] shows the improvement in depicting the shape of the wire insert compared to the 180° mode.
One limitation for this study was the suitability of some of the phantom inserts for assessing the IQ in dental CBCT. For the spatial resolution test, the bar pattern inserts were not able to discriminate between the four voxel sizes, although the MTF calculation did demonstrate a difference, with two resolution options available on this machine. Since dental CBCT machines often reconstruct the volumetric image by binning together adjacent pixels in the projections or by some other means of resampling the data, having a better bar pattern resolution test object, covering a wide range of resolutions intermediate to those already present, would be beneficial. Although the MTF did provide a more accurate estimate of the spatial resolution, additional calculations are needed that require some expertise to perform and interpret and may be less immediate if the expertise is not available on site. Furthermore, as the technology evolves and the spatial resolution improves, a wider range of resolutions may be needed to cover the full range of CBCT machines commercially available. The contrast resolution insert could also be tweaked to include more diagnostically relevant tasks. In the current configuration, all of the test objects are visualized against a PMMA background, which has similar attenuation to soft tissue. Only Delrin exhibited differences in the ability to visualize the test object against the PMMA background because these two materials have similar attenuation properties. However, in dental applications, the object of interest is not necessarily surrounded with soft tissue. By altering some of the test objects to include high-density test objects embedded in a background material with more similar X-ray attenuation properties, the dental applications of visualizing implanted devices, enamel, dentin, or bone could be more accurately tested, as these materials are not surrounded by soft tissue in clinical practice.
| Conclusions|| |
We have measured five IQ metrics and dose in PMMA phantoms for different image acquisition settings on a CBCT machine. Changing the FOV size had the smallest impact on IQ. Using an angular rotation of 180° lead to a reduction in dose to the patient, although small objects were not reconstructed accurately, with artifacts present near all metallic objects. Different voxel size settings did not necessarily lead to improved IQ, small structure conspicuity, or reduced dose to the patient, underlining the need to characterize these parameters for different CBCT models.
The authors wish to thank the Dental Department at the British Columbia Children's Hospital for the use of the iCAT machine and Elham Abouei for the MTF code. Funding for this project was from UBC Faculty of Dentistry Research Equipment Grant (2011).
Financial support and sponsorship
Funding for this project was from UBC Faculty of Dentistry Research Equipment Grant (2011).
Conflicts of interest
There are no conflicts of interest.
| References|| |
Hatcher DC. Operational principles for cone-beam computed tomography. J Am Dent Assoc 2010;141 Suppl 3:3S-6S.
Horner K, Jacobs R, Schulze R. Dental CBCT equipment and performance issues. Radiat Prot Dosimetry 2013;153:212-8.
Liang X, Jacobs R, Hassan B, Li L, Pauwels R, Corpas L, et al.
A comparative evaluation of cone beam computed tomography (CBCT) and multi-slice CT (MSCT) Part I. On subjective image quality. Eur J Radiol 2010;75:265-9.
Loubele M, Maes F, Jacobs R, van Steenberghe D, White SC, Suetens P. Comparative study of image quality for MSCT and CBCT scanners for dentomaxillofacial radiology applications. Radiat Prot Dosimetry 2008;129:222-6.
Suomalainen A, Kiljunen T, Käser Y, Peltola J, Kortesniemi M. Dosimetry and image quality of four dental cone beam computed tomography scanners compared with multislice computed tomography scanners. Dentomaxillofac Radiol 2009;38:367-78.
Pauwels R, Stamatakis H, Manousaridis G, Walker A, Michielsen K, Bosmans H, et al.
Development and applicability of a quality control phantom for dental cone-beam CT. J Appl Clin Med Phys 2011;12:3478.
Bamba J, Araki K, Endo A, Okano T. Image quality assessment of three cone beam CT machines using the SEDENTEXCT CT phantom. Dentomaxillofac Radiol 2013;42:20120445.
Choi JW, Lee SS, Choi SC, Heo MS, Huh KH, Yi WJ, et al.
Relationship between physical factors and subjective image quality of cone-beam computed tomography images according to diagnostic task. Oral Surg Oral Med Oral Pathol Oral Radiol 2015;119:357-65.
Brüllmann D, Schulze RK. Spatial resolution in CBCT machines for dental/maxillofacial applications-what do we know today? Dentomaxillofac Radiol 2015;44:20140204.
Mischkowski RA, Pulsfort R, Ritter L, Neugebauer J, Brochhagen HG, Keeve E, et al.
Geometric accuracy of a newly developed cone-beam device for maxillofacial imaging. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:551-9.
Kim TS, Caruso JM, Christensen H, Torabinejad M. A comparison of cone-beam computed tomography and direct measurement in the examination of the mandibular canal and adjacent structures. J Endod 2010;36:1191-4.
Moshiri M, Scarfe WC, Hilgers ML, Scheetz JP, Silveira AM, Farman AG. Accuracy of linear measurements from imaging plate and lateral cephalometric images derived from cone-beam computed tomography. Am J Orthod Dentofacial Orthop 2007;132:550-60.
Waltrick KB, Nunes de Abreu Junior MJ, Corrêa M, Zastrow MD, Dutra VD. Accuracy of linear measurements and visibility of the mandibular canal of cone-beam computed tomography images with different voxel sizes: An in vitro
study. J Periodontol 2013;84:68-77.
Horner K. Review article: Radiation protection in dental radiology. Br J Radiol 1994;67:1041-9.
American Dental Association Council on Scientific Affairs. The use of cone-beam computed tomography in dentistry: An advisory statement from the American Dental Association Council on Scientific Affairs. J Am Dent Assoc 2012;143:899-902.
Ludlow JB, Ivanovic M. Comparative dosimetry of dental CBCT devices and 64-slice CT for oral and maxillofacial radiology. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;106:106-14.
Roberts JA, Drage NA, Davies J, Thomas DW. Effective dose from cone beam CT examinations in dentistry. Br J Radiol 2009;82:35-40.
Pauwels R, Beinsberger J, Collaert B, Theodorakou C, Rogers J, Walker A, et al.
Effective dose range for dental cone beam computed tomography scanners. Eur J Radiol 2012;81:267-71.
Hirsch E, Wolf U, Heinicke F, Silva MA. Dosimetry of the cone beam computed tomography Veraviewepocs 3D compared with the 3D Accuitomo in different fields of view. Dentomaxillofac Radiol 2008;37:268-73.
Palomo JM, Rao PS, Hans MG. Influence of CBCT exposure conditions on radiation dose. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:773-82.
Deman P, Atwal P, Duzenli C, Thakur Y, Ford NL. Dose measurements for dental cone-beam CT: A comparison with MSCT and panoramic imaging. Phys Med Biol 2014;59:3201-22.
[Figure 8], [Figure 7], [Figure 6], [Figure 5], [Figure 4], [Figure 3], [Figure 2], [Figure 1]
[Table 7], [Table 6], [Table 5], [Table 4], [Table 3], [Table 2], [Table 1]
|This article has been cited by|
||Optimization of cone beam computed tomography image quality in implant dentistry
| ||Yasmine Alawaji,David S. MacDonald,Georgios Giannelis,Nancy L. Ford |
| ||Clinical and Experimental Dental Research. 2018; |
|[Pubmed] | [DOI]|
||Development of a New Cone-Beam Computed Tomography Software for Endodontic Diagnosis
| ||Mike R. Bueno,Carlos Estrela,Brunno C. Azevedo,Anibal Diogenes |
| ||Brazilian Dental Journal. 2018; 29(6): 517 |
|[Pubmed] | [DOI]|
||Density estimation based on the Hounsfield unit value of cone beam computed tomography imaging of the jawbone system
| ||Maya Genisa,Solehuddin Shuib,Zainul Ahmad Rajion,Erry Mochamad Arief,Maman Hermana |
| ||Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine. 2018; 232(12): 1168 |
|[Pubmed] | [DOI]|
||Effect of field-of-view size on gray values derived from cone-beam computed tomography compared with the Hounsfield unit values from multidetector computed tomography scans
| ||Abbas Shokri,Leila Ramezani,Mohsen Bidgoli,Mahdi Akbarzadeh,Karim Ghazikhanlu-Sani,Hamed Fallahi-Sichani |
| ||Imaging Science in Dentistry. 2018; 48(1): 31 |
|[Pubmed] | [DOI]|
||Comparative analysis of prevalence of apical periodontitis in smokers and non-smokers using cone-beam computed tomography
| ||Hanan A. Balto,Lama Alabdulaaly,Shaima Bahammam,Asmaća A. Al-Ekrish |
| ||The Saudi Dental Journal. 2018; |
|[Pubmed] | [DOI]|