![(2918) Enas Antar, et al.E.D.J. Vol. 57, No. 4
of the defect and its response to the regenerative
procedures by monitoring the changes in its size
through out the follow up(3)
.
Manyradiographictechniqueshavebeenusedfor
assessment of linear and volumetric measurements
including plain radiography as periapical(4,5)
, pano-
ramic(6,7)
, and lateral cephalometric radiographs(8)
,
and cross-section imaging as conventional tomogra-
phy(9,10)
, computed tomography (11,12,13)
, and recently
cone beam computed tomography(14,15,16,17,18)
which
is a three dimensional imaging modality having low
dose of radiation, fast scan time, isotropic resolu-
tion of the images, variable field of view, and pos-
sibility of interaction with data using personal com-
puter(19,20)
. As cone beam CT is a relatively recent
three dimensional imaging modality so the accuracy
of linear and volumetric measurements taken from
its images need to be farther investigated for con-
firmation.
Materials and Methods
The study was conducted on four dry human
jaws–two mandibles and two maxillae– obtained
from Department of Anatomy-Faculty of Medicine-
Cairo University, thirty single rooted teeth sockets
were selected free from fenestrations or periapical
defects.
Samples preparation: The sockets were painted
internally with a thin layer of nail varnish, then small
gutta percha (GP) pieces (radiopaque marker) were
attached on the facial and lingual or palatal surfaces
of the dry specimens by means of glue. Each dry
jaw received 21 pieces of GP distributed at seven
selected areas. Three GP pieces were attached in
each area, two on the facial surface, one occlusaly
and one apically, both were on the same line and
perpendicular to the horizontal plane, and one on
the lingual or palatal surface opposing the occlusaly
placed buccal piece. The seven areas were selected
as one in the midline, and on each side one at the
canine region, one at the premolars region, and one
at the second molar region.
Direct linear measurements: Direct linear
measurements on the dry sample were obtained
using a digital caliper IOS-USA®
with an accuracy
of 0.01 mm. at each area three measurements were
taken including:
1) Bone height: as the distance between the
apical end of the occlusally placed GP piece and
the occlusal end of the apically placed GP piece on
the facial surface, 2) Bone thickness: as the distance
from occlusal end of occlusally placed buccal GP
piece to the occlusal end of the GP piece palataly
or lingually placed. 3) Bone width: as the distance
between the occlusal ends of two adjacent GP pieces
on the facial surface. Fig. (1)
Direct volumetric measurements: impressions
for the sockets were taken using [Speedex®
]
putty and light impression material by wash
impression technique. Fig. (2) Then volumes of
these impressions were calculated according to
the water displacement technique as following:
A 10 ml measuring cylinder with an accuracy of
0.1 ml was filled with water at room temperature
to a 6 ml mark. Then each socket impression was
Fig. (1) The digital caliper usage in determination of the bone
height (top), thickness (middle), width (bottom).](https://image.slidesharecdn.com/determinationoftheaccuracyoflinearand-150407015401-conversion-gate01/85/Determination-of-the-accuracy-of-linear-and-volumetric-measurements-on-CBCT-images-2-320.jpg)
![Determination of the Accuracy of Linear and Volumetric (2919)
immersed in the measuring cylinder and the volume
of the displaced water was determined using 0.1ml
measuring pipette (graded with 0.001 ml grades).
The volume of each tooth socket was measured
three times and the average volume was considered
the gold standard.
Sample scanning using CBCT: Each dry
specimen was scanned using i-CAT®
(Imaging
Sciences International, Hatfield, PA) scanner,
at (5mA) , (120 kV), with a single 360˚ rotation
and total scan time of (7 sec), and voxel size of
0.125 mm.
CBCT linear measurements were made on a
personal computer using distance icon on the tool
bar of i-Vision software, the image slices with the
radiopaque markers best visible were used for linear
measurements fig (3).
CBCT volumetric measurements: The CBCT
data set was then exported using the DICOM
(Digital Imaging and Communication in Medicine)
file format to a workstation for automated volume
measurement using separate software SimPlant®
(Materialise dental-Belgium) which allows
segmenting the teeth sockets on consecutive axial
slices as for segmentation a threshold for air was
selected then the region growing tool was applied
which allows selecting the largest connected volume
that contains all voxels with grey values lying inside
the defined threshold, then the selected socket
was displayed in a 3D form and its volume was
calculated using volume calculation function of the
software. These procedures were repeated for each
socket, and each socket in the sample was colour
coded to facilitate differentiation between them.
Statistical analysis: Estimation of the
measurements error between direct (gold standard)
measurements (A) and CBCT measurements (B)
was assessed according to the following equations:
• Absolute measurement error (AME) = [B – A]
• Absolute percentage error (APE) = [B–A] X100 %
A
The paired t test was used for statistical
analysis. Results were considered to be statistically
significant at P ≤ 0 .05. Reliability analysis tests
using Cronbach’s alpha and intra-class correlation
coefficient were used to determine the consistency
between the CBCT and direct measurements.
Higher values of Cronbach’s alpha (More than
0.7) denote acceptable consistency or agreement
between the methods. The significance level was
set at P ≤ 0.05. Statistical analysis was performed
with SPSS 16.0®
(Statistical Package for Scientific
Studies) for Windows.
Fig. (2) Impressions of the sockets for volume assessment of
these sockets.
Fig. (3) Implant screen for skull scan showing simulated
panoramic image of maxilla (top) and seven serial
trans-axial cross-sectional images (bottom), one of the
last showing height and thickness linear measurements
(in blue lines)](https://image.slidesharecdn.com/determinationoftheaccuracyoflinearand-150407015401-conversion-gate01/85/Determination-of-the-accuracy-of-linear-and-volumetric-measurements-on-CBCT-images-3-320.jpg)
Determination of the accuracy of linear and volumetric measurements on CBCT images
- 1. I . S . S . N 0 0 7 0 - 9 4 8 4 w w w . e d a - e g y p t . o r g EGYPTIAN DENTAL JOURNAL Vol. 57, 2917:2925, October, 2011 Determination of the Accuracy of Linear and Volumetric Measurements on Cone Beam Computed Tomography Images (In-vitro study) ABSTRACT Objectives: to determine the accuracy of the linear and volumetric measurements on cone beam computed tomography images. Materials and methods: on four dry human jaws impressions for single rooted teeth sockets were taken then their volumes were determined by “water displacement technique” and compared with their volumes on CBCT scans. On each jaw the linear distances between gutta percha pieces glued at certain positions were measured directly by means of digital caliper, and on CBCT scans of these jaws and then compared. Results: there was no statistically significant difference between direct and CBCT linear and volumetric measurements, for CBCT linear measurements the average AME was 0.06mm and the average error APE was 0.5%. While for CBCT volumetric measurements the average AME was 3.7 mm3 and the averageAPE was 4.9%. Conclusions: there is a positive correlation between (AME) and distance length in CBCT linear measurements, whereas for CBCT volumetric measurements there is an inverse correlation between (APE) and the volume size, CBCT linear measurements are the most accurate at midline region and least accurate at canine region. Keywords: Cone Beam CT, linear measurements, volumetric measurements. Enas Antar* ; Mohamed Khalifa** and Sahar Hosny*** * Instructor, Oral Radiology Dept., Faculty of Oral and Dental Medicine, Cairo University ** Lecturer, Oral Radiology Dept., Faculty of Oral and Dental Medicine, Cairo University *** Associated Professor, Oral Radiology Dept., Faculty of Oral and Dental Medicine, Cairo University INTRODUCTION Nowadays accurate linear and volumetric measurements have become essential in most of the dental branches. As in implant dentistry, pre- surgical dental implant planning requires accurate evaluation of bone in terms of width, height and quality (1) . In diagnosis and treatment planning for orthodontic patients it is necessary to calculate spatial and angular relations between specific landmarks (2) , also In patients with periodontal problems, linear measurements are needed to assess the alveolar bone level and degree of alveolar bone resorption, and in case of intra-bony defect, volumetric measurement is needed to asses the size
- 2. (2918) Enas Antar, et al.E.D.J. Vol. 57, No. 4 of the defect and its response to the regenerative procedures by monitoring the changes in its size through out the follow up(3) . Manyradiographictechniqueshavebeenusedfor assessment of linear and volumetric measurements including plain radiography as periapical(4,5) , pano- ramic(6,7) , and lateral cephalometric radiographs(8) , and cross-section imaging as conventional tomogra- phy(9,10) , computed tomography (11,12,13) , and recently cone beam computed tomography(14,15,16,17,18) which is a three dimensional imaging modality having low dose of radiation, fast scan time, isotropic resolu- tion of the images, variable field of view, and pos- sibility of interaction with data using personal com- puter(19,20) . As cone beam CT is a relatively recent three dimensional imaging modality so the accuracy of linear and volumetric measurements taken from its images need to be farther investigated for con- firmation. Materials and Methods The study was conducted on four dry human jaws–two mandibles and two maxillae– obtained from Department of Anatomy-Faculty of Medicine- Cairo University, thirty single rooted teeth sockets were selected free from fenestrations or periapical defects. Samples preparation: The sockets were painted internally with a thin layer of nail varnish, then small gutta percha (GP) pieces (radiopaque marker) were attached on the facial and lingual or palatal surfaces of the dry specimens by means of glue. Each dry jaw received 21 pieces of GP distributed at seven selected areas. Three GP pieces were attached in each area, two on the facial surface, one occlusaly and one apically, both were on the same line and perpendicular to the horizontal plane, and one on the lingual or palatal surface opposing the occlusaly placed buccal piece. The seven areas were selected as one in the midline, and on each side one at the canine region, one at the premolars region, and one at the second molar region. Direct linear measurements: Direct linear measurements on the dry sample were obtained using a digital caliper IOS-USA® with an accuracy of 0.01 mm. at each area three measurements were taken including: 1) Bone height: as the distance between the apical end of the occlusally placed GP piece and the occlusal end of the apically placed GP piece on the facial surface, 2) Bone thickness: as the distance from occlusal end of occlusally placed buccal GP piece to the occlusal end of the GP piece palataly or lingually placed. 3) Bone width: as the distance between the occlusal ends of two adjacent GP pieces on the facial surface. Fig. (1) Direct volumetric measurements: impressions for the sockets were taken using [Speedex® ] putty and light impression material by wash impression technique. Fig. (2) Then volumes of these impressions were calculated according to the water displacement technique as following: A 10 ml measuring cylinder with an accuracy of 0.1 ml was filled with water at room temperature to a 6 ml mark. Then each socket impression was Fig. (1) The digital caliper usage in determination of the bone height (top), thickness (middle), width (bottom).
- 3. Determination of the Accuracy of Linear and Volumetric (2919) immersed in the measuring cylinder and the volume of the displaced water was determined using 0.1ml measuring pipette (graded with 0.001 ml grades). The volume of each tooth socket was measured three times and the average volume was considered the gold standard. Sample scanning using CBCT: Each dry specimen was scanned using i-CAT® (Imaging Sciences International, Hatfield, PA) scanner, at (5mA) , (120 kV), with a single 360˚ rotation and total scan time of (7 sec), and voxel size of 0.125 mm. CBCT linear measurements were made on a personal computer using distance icon on the tool bar of i-Vision software, the image slices with the radiopaque markers best visible were used for linear measurements fig (3). CBCT volumetric measurements: The CBCT data set was then exported using the DICOM (Digital Imaging and Communication in Medicine) file format to a workstation for automated volume measurement using separate software SimPlant® (Materialise dental-Belgium) which allows segmenting the teeth sockets on consecutive axial slices as for segmentation a threshold for air was selected then the region growing tool was applied which allows selecting the largest connected volume that contains all voxels with grey values lying inside the defined threshold, then the selected socket was displayed in a 3D form and its volume was calculated using volume calculation function of the software. These procedures were repeated for each socket, and each socket in the sample was colour coded to facilitate differentiation between them. Statistical analysis: Estimation of the measurements error between direct (gold standard) measurements (A) and CBCT measurements (B) was assessed according to the following equations: • Absolute measurement error (AME) = [B – A] • Absolute percentage error (APE) = [B–A] X100 % A The paired t test was used for statistical analysis. Results were considered to be statistically significant at P ≤ 0 .05. Reliability analysis tests using Cronbach’s alpha and intra-class correlation coefficient were used to determine the consistency between the CBCT and direct measurements. Higher values of Cronbach’s alpha (More than 0.7) denote acceptable consistency or agreement between the methods. The significance level was set at P ≤ 0.05. Statistical analysis was performed with SPSS 16.0® (Statistical Package for Scientific Studies) for Windows. Fig. (2) Impressions of the sockets for volume assessment of these sockets. Fig. (3) Implant screen for skull scan showing simulated panoramic image of maxilla (top) and seven serial trans-axial cross-sectional images (bottom), one of the last showing height and thickness linear measurements (in blue lines)
- 4. (2920) Enas Antar, et al.E.D.J. Vol. 57, No. 4 5. Height measurements Measurement Min Max Mean SD Direct 5.25 mm 15.74 mm 10.09 mm 2.92 mm CBCT 5.25 mm 15.74 mm 10.07 mm 2.91 mm AME 0.00 mm 0.23 mm 0.06 mm 0.038 mm APE 0% 1.7% 0.5% 0.36% 6. Thickness linear measurements Measurement Min Max Mean SD Direct 4.19 mm 14.31 mm 8.30 mm 2.84 mm CBCT 4.20 mm 14.32 mm 8.31 mm 2.85 mm AME 0.00 mm 0.16 mm 0.045 mm 0.042 mm APE 0% 1.6% 0.48% 0.43% 7. Width linear measurements Measurement Min Max Mean SD Direct 13.17 mm 23.95 mm 17.75 mm 3.09 mm CBCT 13.17 mm 23.92 mm 17.76 mm 3.10 mm AME 0.00 mm 0.33 mm 0.09 mm 0.08 mm APE 0% 1.5% 0.48% 0.42% 8. All linear measurements Measurement Min Max Mean SD Direct 4.19 mm 23.95 mm 11.76 mm 4.95 mm CBCT 4.20 mm 23.92 mm 11.77 mm 4.96 mm AME 0.00 mm 0.33 mm 0.06 mm 0.054 mm APE 0% 1.7% 0.5% 0.4% Results I. Linear Measurements Table (1) The descriptive statistics for the whole CBCT and direct linear measurements, AME, and APE 1. Midline region linear measurements Measurement Min Max Mean SD Direct 4.19 mm 19.45 mm 12.12 mm 5.28 mm CBCT 4.20 mm 19.40 mm 12.13 mm 5.27 mm AME 0.00 mm 0.15 mm 0.04 mm 0.043 mm APE 0% 1.05% 0.34% 0.31% 2. Canine region linear measurements Measurement Min Max Mean SD Direct 5.26 mm 21.33 mm 12.61 mm 5.91 mm CBCT 5.26 mm 21.46 mm 12.62 mm 5.92 mm AME 0.00 mm 0.31 mm 0.08 mm 0.084 mm APE 0% 1.55% 0.60 % 0.43 % 3. Premolars region linear measurements Measurement Min Max Mean SD Direct 5.32 mm 23.95 mm 12.24 mm 4.66 mm CBCT 5.35 mm 23.86 mm 12.21 mm 4.65 mm AME 0.00 mm 0.16 mm 0.06 mm 0.04 mm APE 0% 1.68% 0.54% 0.37% 4. Molars region linear measurements Measurement Min Max Mean SD Direct 5.64 mm 14.32 mm 9.41 mm 2.68 mm CBCT 5.64 mm 14.31 mm 9.44 mm 2.68 mm AME 0.00 mm 0.16 mm 0.05 mm 0.04 mm APE 0% 1.68% 0.54% 0.37%
- 5. Determination of the Accuracy of Linear and Volumetric (2921) II. Volumetric Measurements Table (2) The descriptive statistics for the whole CBCT and direct linear measurements, AME, and APE 1. Maxillary volumetric measurements Measurement Min Max Mean SD Direct 60 mm3 271 mm3 141.79 mm3 74.06 mm3 CBCT 65 mm3 275 mm3 144.79 mm3 73.23 mm3 AME 2.5 mm3 10 mm3 5.8 mm3 2.4 mm3 APE 1.8% 10% 4.46% 2.4% 2. Mandibular volumetric measurements Measurement Min Max Mean SD Direct 42 mm3 140 mm3 71.80 mm3 29.21 mm3 CBCT 39.50 mm3 137.50 mm3 70.50 mm3 30.04 mm3 AME 0.5 mm3 6 mm3 3.4 mm3 1.52 mm3 APE 0.83% 9.6% 5.4% 2.8% 3. All volumetric measurements Measurement Min Max Mean SD Direct 42 mm3 271 mm3 105.6 mm3 65.1 mm3 CBCT 39.5 mm3 275 mm3 106.3 mm3 66.1 mm3 AME 0.5 mm3 10 mm3 3.7 mm3 2.3 mm3 APE 0.83% 10% 4.9% 2.6% CBCT volumetric measurements showed underestimation in 17 sockets ranged from (0.5mm3 to 10mm3 ) and overestimation in 13 sockets ranged from (1mm3 to 9mm3 ). Paired-t test results for comparison between the volumetric measurements obtained by direct and CBCT methods showed that there was no statistically significant difference between the two groups. Also, Cronbach’s alpha and intra-class correlation coefficient values showed statistically significant agreement between the CBCT and direct volumetric measurements with a very good reliability coefficient. From 80 linear measurements CBCT showed underestimation in 29 measurements ranged from (0.01 to 0.33mm) and overestimation in 42 measurements ranged from (0.01 to 0.31 mm). Using paired t test analysis there was no statistically significant difference between the direct and CBCT linear measurements, Cronbach’s alpha and intra-class correlation coefficient values showed statistically significant agreement between the CBCT and direct linear measurements with a perfect reliability coefficient Paired-t test results for comparison between maxillary and mandibular CBCT linear measurements (AME) and (APE) showed that there was no statistically significant difference between the two groups in both AME and APE One Way Analysis of Variance (ANOVA) test was used for comparison between theAME andAPE for CBCT linear measurements at different regions of the jaw, the significance level was set at P ≤ 0.05. The results showed that there was no statistically significant difference between the different regions in both AME and APE One Way Analysis of Variance (ANOVA) test was used for comparison between the averages of AME and APE for CBCT linear measurements in different dimensions, the significance level was set at P ≤ 0.05. The results showed that there was no statistically significant difference between the different dimensions in APE while for AME the results showed that there was statistically significant difference. Pair-wise Multiple Comparison Procedures (Holm-Sidak method) showed the presence of statistically significant difference between width and thickness AME; nevertheless no statistically significant difference was found between AME of height neither with width nor with thickness
- 6. (2922) Enas Antar, et al.E.D.J. Vol. 57, No. 4 (Mercury Computer Systems, Chelmsford, MA). For assessment of accuracy of linear measurements seven areas were selected on each jaw including anterior, canine, premolar, and molar regions. The GP pieces used were of small size to avoid possible production of streaking artifacts in CBCT images. The CBCT linear measurements in this study were proved to be accurate and highly reliable although they were not constant in one direction (overestimation or underestimation) but they showed a higher tendency to overestimation. Comparison between CBCT height, thickness and width measurements using Pair-wise Multiple Comparison Procedures (Holm-Sidak method) showed the presence of a statistically significant difference between width and thickness AME; nevertheless no statistically significant difference was found between AME of height neither with width nor with thickness. However the results of ANOVA test for comparison between the APE at different dimensions showed no statistically significant difference. However this difference can be explained as the linear distances that represent the width were larger than the linear distances that represent the height and thickness as the average real width measurements was 17.75 mm, while those for thickness and height respectively were 8.30 mm and 10.9 mm respectively. This coincided with (Mischkowski et al 2007) who stated that “there is a positive correlation between AME and distance length in CBCT-based measurements”. The CBCT linear measurements accuracy of this study was higher than CBCT linear measure- ments accuracy of Kobayashi et al 2004 who had examined the accuracy of linear measurements on limited cone beam CT scans with 0.117 voxel reso- lution the average AME for was 0.22 mm and the average APE was 1.4% which are higher than those Paired-t test results for comparison between maxillary and mandibular CBCT volumetric measurements (AME) and (APE) showed that there was no statistically significant difference between the two groups in APE, but there was statistically significant difference between the two groups in AME Discussion The sockets selected for assessment of accuracy of CBCT volumetric measurements were single rooted, free from periapical defects and fenestration to facilitate impression taking with minimum error and not to invite errors in automated volume extraction, the sockets were painted internally to block the minute openings of the cancellous bone which both facilitated impression removal and made delineation of the sockets walls in the CBCT images thus facilitating and giving more accurate automatic volume extraction. The impressions of the sockets were taken using condensation silicon rubber base impression material having a very high degree of accuracy in the first hour before polymerization shrinkage begins(21) . The volumes of sockets’ impressions were measured by water displacement technique which was proved to be accurate in determining the objects volume and it had been used by Emirzeoglu et al 2005, Shiratori et al 2005 and by Yang et al 2006 in determination of different objects volumes. The sockets’ volumes on the CBCT images were measured by automatic volume extraction function of Simplant® software for rapid, ease, and accuracy of the procedures when compared with manual delineation of the socket outline on each socket segment which is time consuming and inviting humane error. The automated volume extraction technique was used by Pinsky et al 2006 utilizing software called Analyze® (Analyze Direct Inc., MN), and by Mischkawski et al 2007 utilizing software called Amira 3.1.1 visualization®
- 7. Determination of the Accuracy of Linear and Volumetric (2923) between maxillary and mandibular CBCT linear measurements accuracy, AL-Ekrish and Ekram 2011 found a statistically significant difference between them in AME, with a higher AME for maxillae than that for mandibles, while in the current study no statistically significant difference was observed between maxillary and mandibular linear measurements but there was a slight higher accuracy for mandibular measurements (as in their study). AL-Ekrish and Ekram 2011 also found no difference between the different sites in CBCT linear measurements accuracy, whereas in the present study although comparison between different regions of the jaw in CBCT linear measurements accuracy using One Way Analysis of Variance (ANOVA) test showed no statistically significant difference between the different regions in both AME and APE, but regarding the average values ofAME andAPE the highest accuracy was at midline linear measurements, and the least accuracy was at the canine region linear measurements, while for premolars and molars regions the averages AME were very close and the averages APE were the same. All the regions showed no statistically significant difference between CBCT and direct measurements with Paired-t test, and statistically significant agreement between CBCT and direct measurements with reliability analysis. In this study we also evaluated the accuracy of CBCT volumetric measurements using a volume measurement approach from those offered by popular softwares used for computer-aided surgical procedures. CBCT volumetric measurements were proved to be accurate and reliable, they were not consistent in one direction (overestimation or underestimation) as they showed underestimation in 17 sockets and overestimation in 13 sockets, and the AME values in both directions were very close. However the mandibular CBCT volumetric measurements showed a greater tendency for underestimation. of our study this could be due to the presence of the soft tissue on the cadavers’ mandibles which most probably negatively influenced the image quality of the scans, and in turn the measurements accuracy. While by comparison with the results of Pinsky et al 2006 it was revealed that they had a mean width (diameter) error of 0.01mm and mean height (depth) error of 0.03mm for the acrylic block– (both were lower than the AME in this study)– while for the human mandible the mean width (diameter) error was 0.07 mm–(which was almost equal to that in this study) – and the mean height (depth) error was 0.27 mm (which was much higher than the AME in this study). This difference can be explained on the basis of the dry mandible wide trabeculations and various bone densities which lead to variations in CBCT densities which could impact the results. The averages AME and APE for CBCT linear measurements in Mischkowski et al 2007 study were 0.26mm and 0.98% which are greater than those of this study Anyway the distances measured in Mischkowski’s study were greater than the distances in this study, and by using linear regression analysis they concluded that there was a positive correlation between AME and distance length in CBCT-based measurements which is in agreement with this study results. AME in Berco et al 2009 study is almost similar or very close to that in our study, while the APE of their study was in range of 0–1.1% with an average of 0.19% which is smaller than that in our study, this difference exist because Berco et al 2009 disagreed with our and Mischkowski et al 2007 conclusion that there is a positive correlation between AME and distance length in CBCT-based measurements. In contrast to the current study and all the previously mentioned studies; AL-Ekrish and Ekram 2011 found that both MDCT and CBCT were associated with clinically and statistically significant measurement error. In comparison
- 8. (2924) Enas Antar, et al.E.D.J. Vol. 57, No. 4 • The CBCT linear measurements are the most accurate at midline region and least accurate at canine region. • There is a positive correlation between (AME) and distance length in CBCT-based linear measurements • CBCT is accurate and reliable for volumetric measurements • CBCT volumetric measurements are not con- stant in one direction (overestimation or under- estimation) but they have a higher tendency to underestimation in small sized volumes • There is an inverse correlation between (APE) and the volume size in CBCT-based volumetric measurements References 1) Lofthag-Hansen S, Gröndahl K, Ekestubbe A. Cone-Beam CT for Preoperative Implant Planning in the Posterior Mandible: Visibility of Anatomic Landmarks. Clin Im- plant Dent Relat Res 11:246-255, 2009. 2) Berco M, Rigali P H, Matthew Miner R, Deluca S, An- dreson NK, Will LA. Accuracy and reliability of linear cephalometric measurements from cone-beam computed tomography scans of a dry human skull. Am J Orthod Den- tofacial Orthop 2009; 136:17.e1-17.eg. 3) Grimard BA, Hoidal MJ, Mills MP, Mellonig JT, Num- mikoski PV, Mealey BL. Comparison of clinical, peri- apical radiograph, and cone-beam volume tomography measurement techniques for assessing bone level changes following regenerative periodontal therapy. J Periodontol. 2009 Jan; 80 (1):48-55. 4) Ellingsen MA, Hollener LG, Harrington GW. Radiovisi- ography versus conventional radiography for detection of small instruments in endodontic length determination. II. In vivo evaluation. J Endod. 1995 Oct; 21 (10):516-20. 5) Sanabe ME, Basso MD, Goncalves MA, Cordeiro RCL, Santos-Pinto LD. Digital versus conventional radiography for determination of primary incisor length. Braz J Oral Sci. April/June 2009. 8(2): 101-4. 6) Chuenchompoonut V, Ida M, Honda E, Kurabayashi T, Sasaki T. Accuracy of panoramic radiography in assess- Comparing the accuracy of maxillary and man- dibular CBCT volumetric measurements showed no a statistically significant difference between the two groups in APE, but there was a statistically signifi- cant difference between both groups in AME. Correlating this difference with the fact that the real maxillary sockets’ volume average (141.79 ± 74.06 mm3 ) was greater than the real mandibular sockets volume average (71.8 ± 29.2 mm3 ) put our study in agreement with Winer-Muram et al 2003, and Goo et al 2005 they concluded that there is an inverse correlation between measurement error percentage (APE) and the volume size. Comparing the results of Pinsky et al 2006 study utilizing the automated volume extraction with the results of this study revealed that this study had a smaller AME but a greater APE which could be due to the conclusion we had of the inverse relation between APE and the volume size. The APE of our study was lesser than APE of Yang et al 2006 study (7.6%) most probably this because the manual step in volume calculation in Yang’s study which invited human error. The average APE for CBCT examination in Mischkowski et al 2007 was 6.01% which is greater than the APE of CBCT volumetric measurements in this study. Conclusions • Cone beam computed tomography is accurate and highly reliable in linear measurements • Although CBCT linear measurements are not constant in one direction (overestimation or underestimation) but they have a higher tendency to overestimation, and the AME values in both directions are very close. • The accuracy of CBCT linear measurements in mandibles is nearly the same as or slightly higher than in maxillae
- 9. Determination of the Accuracy of Linear and Volumetric (2925) 15) Mischkowski RA, Pulsfort R, Ritter L, Neugebauer J, Bro- chhagen HG, Kevee E, Zöller JE. 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. 16) Pinsky HM, Dyda S, Pinsky RW, Misch KA, Sarment DP. Accuracy of three dimensional measurements using cone- beam CT. Dentomaxillofac Radiol 35:410-416, 2006. 17) Gracco A, Lombardo L, Cozzani M, Siciliani G. Quantita- tive cone-beam computed tomography evaluation of palatal bone thickness for orthodontic mini-screw placement. Am J Orthod Dentofacial Orthop. 2008 Sep; 134 (3): 361-9. 18) Quereshy A, Savell A, Palomo M. Applications of cone beam computed tomography in the practice of oral and maxillofa- cial surgery. J Oral Maxillofac Surg. 66:791-796, 2008 19) Scarfe WC, Farman AG, Sukovic P. Clinical applications of cone-beam computed tomography in dental practice. J Can Dent Assoc 72:75-80, 2006. 20) Scarfe WC, Farman AG. What is cone-beam CT and how does it work? Dent Clin North Am 52:707-30, v., 2008. 21) Giordano R. Impression materials: basic properties. Gen Dent 2000;48:510 516. 22) Shiratori CA, Schellini SA, Yamashita S, Padovani CR, Rossa R. Rabbit orbital measure evaluations after enocula- tion and evisceration. Arq Bras Oftalmol. 2005 Mar-Apr; 68(2): 235-9. 23) AL-Ekrish AA and Ekram M. A comparative study of the accuracy and reliability of multi-detector computed tom- ography and cone beam computed tomography in the as- sessment of dental implant site dimensions. Dentomaxil- lofacial Radiology 2011. 40, 67-75. 24) Yang F, Jacobs R, Willems G. Dental age estimation through volume matching of teeth imaged by cone beam CT. Fo- rensic Sci Int. 2006 May 15; 159 suppl 1:s78-83. 25) Winer-Muram HT, Jennings SG, Meyer CA, Liang Y, Aisen AM, Tarver RD, et al. Effect of varying CT section width on volumetric measurement of lung tumors and application of compensatory equations. Radiology 2003; 229:184-94. ing the dimensions of radiolucent jaw lesions with distinct or indistinct borders. Dentomaxillofac Radiol. 2003 Mar; 32(2): 80-6. 7) Batenburg RHK, Stellingsma K, Raghoebar GM, Vissink A. Bone height measurements on panoramic radiographs: the effect of shape and position of edentulous mandibles. Oral Surg, Oral Med, Oral Pathol, Oral Radiol, Endod 1997; 84: 430–435. 8) Baumrind S, Frantz RC. The reliability of head film meas- urements. 1. Landmarks identification. Am J Orthod 1971; 60:111-27. 9) Ismail YH, Azarbal M, Kapa SF. Conventional linear tomog- raphy: protocol for assessing endosseous implant sites. J Prosthet Dent 1995; 73:153-7. 10) Silverstien LH, Melkonian RW, Kurtzman D, Garnick JJ, Lefkove MD. Linear tomography in conjunction with pan- tomography in the assessment of dental implant recipient sites. J Oral Implant 1994; 20:111-7. 11) Emirzeoglu M, Sahin B, Selcuk MB, Kaplan S. The effect of section thickness on the estimation of liver volume by the Cavalieri principal using computed tomography im- ages. Eur J Radiol 2005; 56: 391-397. 12) Goo JM, Tongdee T, Tongdee R, Yeo K, Hildebolt CF, Bae KT. Volumetric measurements of synthetic lung noduled with multi-detector row CT: Effect of various image reconstruction parameters and segmentation thresholds on measurements ac- curacy. Radiology. 2005 Jun; 235(3):850-6. 13) Suomalainen A, Vehmas T, Kortesniemi M, Robinson S, Peltola J. Accuracy of linear measurements using dental cone beam and conventional multislice computed tomog- raphy. Dentomaxillofac Radiol. 2008 Jan; 37 (1): 10-7. 14) Kobayashi K, Shlmoda S, Nakawaga Y, Yamamoto A. Ac- curacy in measurement of distance using limited cone- beam computerized tomography. Oral Maxillofac Implants 2004; 19: 228-231.