Abstract
Objectives:
Different factors can affect the discrepancy between the gray value (GV) measurements obtained from CBCT and the Hounsfield unit (HU) derived from multidetector CT (MDCT), which is considered the gold-standard density scale. This study aimed to explore the impact of region of interest (ROI) location and field of view (FOV) size on the difference between these two scales as a potential source of error.
Methods
Three phantoms, each consisting of a water-filled plastic bin containing a dry dentate human skull, were prepared. CBCT scans were conducted using the NewTom VGi evo system, while MDCT scans were performed using Philips system. Three different FOV sizes (8 × 8 cm, 8 × 12 cm, and 12 × 15 cm) were used, and the GVs obtained from eight distinct ROIs were compared with the HUs from the MDCT scans. The ROIs included dental and bony regions within the anterior and posterior areas of both jaws. Statistical analyses were performed using SPSS v. 26.
Results
The GVs derived from CBCT images were significantly influenced by both ROI location and FOV size (p < 0.05 for both factors). Following the comparison between GVs and HUs, the anterior mandibular bone ROI represented the minimum error, while the posterior mandibular teeth exhibited the maximum error. Moreover, the 8 × 8 cm and 12 × 15 cm FOVs resulted in the lowest and highest degrees of GV error, respectively.
Conclusions
The ROI location and the FOV size can significantly affect the GVs obtained from CBCT images. It is not recommended to use the GV scale within the posterior mandibular teeth region due to the potential for error. Additionally, selecting smaller FOV sizes, such as 8 × 8 cm, can provide GVs closer to the gold-standard numbers.
Keywords: Cone-Beam Computed Tomography, Diagnostic Imaging, Multidetector Computed Tomography, Bone Density, Radiographic Phantom
Introduction
CBCT has become a widely used imaging modality in dentistry, offering advantages such as improved quality, three-dimensional visualization, reduced time, accurate sizing of objects, lower radiation dose, and cost-effectiveness compared to multidetector CT (MDCT). 1–5 However, CBCT also has limitations, including limited visualization of soft tissues, higher exposure compared to intraoral and panoramic radiographs, limited assessment of bone density, and the presence of artifacts caused by metal objects. 5,6 CBCT finds extensive applications in oral and maxillofacial surgery, endodontics, implantology, and orthodontics. 1
Assessing the density and quality of the tissue is an important aspect of tomographic imaging, particularly in the context of dental implant placement. 1 When using MDCT, the tissue density is represented by the Hounsfield unit (HU), where water is assigned a value of zero HU, and denser tissues have higher values. 7 In CBCT, the corresponding scale for bone density measurement is the gray value (GV), which has been found to be highly correlated with HU in multiple studies. 8–10 However, there are intrinsic factors related to CBCT that prevent a direct equivalence between these two scales, including the type of imaging device, signal-to-noise ratio, presence of artifacts, acquisition settings, the position of the object in relation to the field of view (FOV), and the size of FOV. 11–19
Parsa et al 17 conducted a study that demonstrated the significant impact of object location on GV within five different FOVs. In contrast, Lagravère et al 20 concluded that the object’s position does not affect the linear relationship between material density and HU. Shokri et al, 18 in another study, found that the difference between GV obtained from CBCT and the HU of MDCT is negligible within a small FOV. However, this difference becomes statistically significant as the FOV size increases. On the other hand, Katsumata et al 21 preferred larger-volume CBCT scans to ensure more consistent density measurements. Thus, ongoing debates exist regarding how FOV size and object’s position can impact the resulting GV in CBCT evaluations.
The objective of this study was to assess how the location of region of interest (ROI) and the FOV size affect the GVs obtained from CBCT. Furthermore, the GVs were compared to the HUs obtained from MDCT, which serves as the gold-standard scale, in order to determine the potential errors that may arise under different circumstances.
Methods and materials
Ethical approval for this study was obtained from the research committee of Shahid Beheshti University of Medical Sciences, Tehran, Iran (IR.SBMU.DRC.REC.1399.082).
Phantom preparation
In this in-vitro study, three phantoms were prepared using dry dentate human skulls devoid of metal objects. The skulls were placed in cylindrical plastic containers filled with distilled water (Figure 1). The containers had a diameter of 16 cm and a height of 17.7 cm. The water depth ranged from 10 to 40 mm in different areas of the container. The skulls were not identified by age, gender, and race, and had to contain intact dental and alveolar posterior and anterior regions in both jaws, preferably.
Figure 1.
The phantom containing dry human skull and filled with distilled water.
Scanning procedure
CBCT imaging was performed using the NewTom VGi evo system (QR, Verona, Italy) with the following settings: voltage of 110 kVp, voxel size of 300 µm, current ranging from 1 to 6 mA, and standard resolution. Prior to scanning, the phantoms were positioned on a plastic surface according to the guiding lines of the device (Figure 2). The three intended FOVs, which contained both jaws completely, were as follows: 8 × 8 cm, 8 × 12 cm, and 12 × 15 cm. Following the power analysis, a minimum of 45 scans was determined to be sufficient for this study. To ensure higher accuracy and validity, each scan was repeated five times with minor repositioning. However, one of the included skulls did not have anterior maxillary teeth ROI, resulting in a total of 345 distinct GV measurements instead of 360. The CBCT scans were converted to Digital Imaging and Communication in Medicine (DICOM) format and transferred to OnDemand 3D software (CyberMed Inc., Seoul, Korea). The same phantoms were subjected to MDCT scanning using 128-slice Philips system (Philips Medical Systems, Netherlands) with a voltage of 200 kVp and a current of 400 mA (Figure 3). MDCT scanning was performed once for each skull, resulting in three gold-standard images. The OnDemand 3D software (CyberMed Inc., Seoul, Korea) was implemented for the GV and HU measurements. GV was calculated for each of the eight ROIs within each scan and compared to the corresponding gold-standard HU obtained from MDCT images.
Figure 2.
The positioning of phantom during the CBCT using NewTom VGi evo device.
Figure 3.
The positioning of phantom during the MDCT scan using Philips device. MDCT, multidetector CT.
Density measurements
The GV was measured by selecting an 8 × 8 square within eight different oral and maxillofacial ROIs, including anterior and posterior dental and bony regions of the maxilla and mandible (Table 1). The regions were determined based on a previous study by Santaella et al. 22 For orientation calibration, the axial sections of scans were synchronized using three anatomic landmarks, including the incisal edge of the left maxillary canine and the cusp tip of the left and right maxillary first molars. The ROIs were accurately assessed using 2 mm thick axial, coronal, and sagittal sections, with the conclusive measurement taken from the axial section.
Table 1.
Comparison of density measurements through two scales of GV and HU within different ROIs
| ROI | Density scale | Sample size | Mean (±SD) | 95% CI | Mean difference | Siga | |
|---|---|---|---|---|---|---|---|
| Lower limit | Upper limit | ||||||
| Anterior mandibular bone | GV | 45 | 1157.73 (±237.82) | 81.51 | 224.41 | 152.96 | 0.000 |
| HU | 3 | 1004.77 (±184.92) | |||||
| Posterior nandibular bone | GV | 45 | 1040.63 (±479.96) | 31.13 | 319.52 | 175.33 | 0.02 |
| HU | 3 | 865.30 (±545.39) | |||||
| Posterior maxillary bone | GV | 45 | 858.01 (±447.65) | 67.86 | 336.84 | 202.35 | 0.004 |
| HU | 3 | 655.66 (±579.40) | |||||
| Anterior maxillary bone | GV | 45 | 888.85 (±276.48) | 128.25 | 294.38 | 211.31 | 0.000 |
| HU | 3 | 677.53 (±223.15) | |||||
| Posterior maxillary teeth | GV | 45 | 1175.97 (±319.56) | -335.55 | -141.15 | -238.35 | 0.000 |
| HU | 3 | 1401.73 (±187.16) | |||||
| Anterior mandibular teeth | GV | 45 | 1330.62 (±312.47) | 235.94 | 423.70 | 329.82 | 0.000 |
| HU | 3 | 1000.80 (±165.42) | |||||
| Anterior maxillary teeth | GV | 30 | 1529.07 (±267.73) | 234.64 | 434.59 | 334.62 | 0.000 |
| HU | 2 | 1194.45 (±96.66) | |||||
| Posterior mandibular teeth | GV | 45 | 1925.30 (±289.98) | 302.98 | 477.22 | 339.10 | 0.000 |
| HU | 3 | 1535.20 (±229.34) | |||||
CI = confidence interval, GV = gray value, HU = Hounsfield unit;ROI = region of interest, Sig = significance.
ROIs are sorted in a descending order of mean difference.
One-sample t-test.
For maxillary anterior teeth ROI definition, the cementoenamel junction (CEJ) of the central incisors was selected within each of the three sections. Next, the distance from the coronal height of contour (HOC) to the pulpal area was calculated, and an 8 × 8 square was illustrated for GV measurement. Regarding the maxillary anterior bone ROI, the interroot distance of canines was measured in the first axial section where the roots of central incisors were not detectable. Similar to anterior teeth, maxillary posterior teeth ROI was defined using an 8 × 8 square near the CEJ of the first molar, within the distance of coronal buccal HOC to pulpal radiolucency. For instance, Figure 4 visualizes the ROI definition for maxillary anterior teeth and bone. The first axial section where the first molars’ roots were not noticeable was used for the posterior maxillary bone ROI. Then, the square was outlined at the midpoint between the buccal and palatal cortical plates of the bone. The same procedures were applied to define the ROIs for mandibular dental and bony regions. The mean (±SD), minimum, and maximum GVs were recorded for each ROI. All evaluations were performed by a dental student under the supervision of an expert oral and maxillofacial radiologist.
Figure 4.
(a) The method used for selecting the anterior maxillary teeth ROI. The CEJ of central incisor was detected and the distance from coronal HOC to pulpal radiolucency was measured. A 8 × 8 square was outlined within this distance. (b) The method used for selecting the anterior maxillary bone ROI. The distance between the roots of canines was calculated in the most inferior axial section with no traceable roots of central incisors. A similar square was selected. CEJ, cementoenamel junction; HOC, height of contour; ROI, region of interest.
Calibration
To assess intrarater reliability, the interclass correlation coefficient (ICC) test was performed by double-assessing 20 regions with a 2-week interval.
Statistics
Descriptive data were presented using tables, diagrams, and statistical parameters. Kruskal–Wallis analyses were conducted to assess the association between GV, FOV size, and ROI location in CBCT. One- and two-sample t-tests were utilized to compare GV with HU concerning different FOV sizes and the ROI locations. All analyses were conducted using SPSS v. 26 (IBM Corp., Armonk, NY) for Windows, and statistical significance was set at a p-value of less than 0.05.
Results
The ICC test demonstrated a high intrarater reliability of 81–95%. Table 1 provides the GVs and HUs categorized by ROI location. The highest mean GV was recorded in the posterior mandibular teeth ROI, while the lowest was measured in the posterior maxillary bone ROI. Regarding HU, posterior mandibular teeth showed the highest amount, while the posterior maxillary bone exhibited the lowest. One-sample t-tests were used to compare GV and HU within different ROIs, revealing significant differences in all regions. The anterior mandibular bone ROI resulted in the least difference and, accordingly, the least error (p < 0.05). The largest difference was observed in posterior mandibular teeth ROI, making this region the most potentially erroneous (p < 0.05).
Due to the non-normal data distribution, the non-parametric Kruskal–Wallis test was used to evaluate the association between GV and ROI location. As shown in Table 2, the ROI location significantly influenced GV (p < 0.05). Pairwise comparisons of GVs within different ROIs were made using Bonferroni correction, which indicated a significant difference in 15 out of 28 comparisons (Table 3).
Table 2.
The influence of ROI location on the gray values obtained from CBCT scans
| ROI | Mean rank | χ2 | Siga |
|---|---|---|---|
| Anterior maxillary bone | 96.98 | 161.34 | 0.000 |
| Anterior maxillary teeth | 244.90 | ||
| Posterior maxillary bone | 97.46 | ||
| Posterior maxillary teeth | 164.23 | ||
| Anterior mandibular bone | 159.27 | ||
| Anterior mandibular teeth | 198.00 | ||
| Posterior mandibular bone | 137.76 | ||
| Posterior mandibular teeth | 309.38 |
ROI = region of interest.
Kruskal–Wallis test.
Table 3.
Pairwise comparison of different regions of interest regarding gray values obtained from CBCT scanning
| Comparison | Statistic | Siga | |
|---|---|---|---|
| Group 1 | Group 2 | ||
| Posterior maxillary bone | Anterior maxillary bone | 0.48 | 1.00 |
| Posterior mandibular bone | Anterior maxillary bone | 40.78 | 1.00 |
| Anterior mandibular bone | Anterior maxillary bone | 62.29 | 0.085 |
| Posterior maxillary teeth | Anterior maxillary bone | 67.26 | 0.039 |
| Anterior mandibular teeth | Anterior maxillary bone | 101.02 | 0.000 |
| Anterior maxillary teeth | Anterior maxillary bone | 147.92 | 0.000 |
| Posterior mandibular teeth | Anterior maxillary bone | 212.40 | 0.000 |
| Posterior mandibular bone | Posterior mmaxillary bone | 40.30 | 1.00 |
| Anterior mandibular bone | Posterior maxillary bone | 61.81 | 0.092 |
| Posterior maxillary teeth | Posterior maxillary bone | 66.78 | 0.042 |
| Anterior mandibular teeth | Posterior maxillary bone | 100.54 | 0.000 |
| Posterior maxillary bone | Anterior maxillary teeth | 147.44 | 0.000 |
| Posterior mandibular teeth | Posterior maxillary bone | 211.92 | 0.000 |
| Posterior mandibular bone | Anterior mandibular bone | 21.51 | 1.00 |
| Posterior mandibular bone | Posterior maxillary teeth | 26.48 | 1.00 |
| Posterior mandibular bone | Anterior mandibular teeth | 60.24 | 0.117 |
| Posterior mandibular bone | Anterior maxillary teeth | 107.14 | 0.000 |
| Posterior mandibular teeth | Posterior mandibular bone | 171.62 | 0.000 |
| Anterior mandibular bone | Posterior maxillary teeth | 4.97 | 1.00 |
| Anterior mandibular yeeth | Anterior mandibular bone | 38.73 | 1.00 |
| Anterior mandibular bone | Anterior maxillary teeth | 85.63 | 0.000 |
| Posterior mandibular teeth | Anterior mandibular bone | 150.11 | 0.000 |
| Anterior mandibular teeth | Posterior maxillary teeth | 33.77 | 1.00 |
| Posterior maxillary teeth | Anterior maxillary teeth | 80.67 | 0.017 |
| Posterior mandibular teeth | Posterior maxillary teeth | 145.14 | 0.000 |
| Anterior mandibular teeth | Anterior maxillary teeth | 46.90 | 1.00 |
| Posterior mandibular teeth | Anterior mandibular teeth | 111.38 | 0.000 |
| Posterior mandibular teeth | Anterior maxillary teeth | 64.48 | 0.171 |
Sig = significance.
Multiple comparison (Bonferroni correction).
Table 4 presents the GV measurements for each FOV, with the highest mean values reported for the 12 × 15 cm FOV and the lowest for the 8 × 8 cm FOV. According to the Kruskal–Wallis test, FOV dimensions significantly impacted GV (p < 0.05). Bonferroni correction confirmed significant differences in all three pairwise comparisons: 8 × 8 cm vs 8 × 12 cm (p < 0.05), 8 × 8 cm vs 12 × 15 cm (p < 0.05), and 12 × 15 cm vs 8 × 12 cm (p < 0.05).
Table 4.
The influence of FOV size on GVs obtained from CBCT scanning
| FOV (cm) | Sample size | GV | Mean rank | Siga |
|---|---|---|---|---|
| Mean (±SD) | ||||
| 8 × 8 | 115 | 1054.37 (±451.60) | 136.78 | 0.000 |
| 8 × 12 | 115 | 1217.23 (±453.25) | 170.62 | |
| 12 × 15 | 115 | 1405.29 (±458.27) | 211.59 | |
| Total | 345 | |||
FOV = field of view, GV = gray value, Sig = significance.
Kruskal–Wallis test.
The effect of FOV size and ROI location on the difference between GV and HU (gold-standard) was simultaneously compared using one- or two-sample t-tests, depending on the normal or non-normal data distribution. Accordingly, the most and the least efficient FOVs within each ROI can be found in Tables 5 and 6. It is depicted that the 12 × 15 cm FOV within the posterior mandibular teeth ROI resulted in the highest mean GV. In contrast, the 8 × 8 cm FOV in the posterior maxillary bone ROI showed the least average. On the whole, the 8 × 8 cm FOV resulted in the minimum error, while the 12 × 15 cm FOV showed the maximum difference from the standard value.
Table 5.
Comparing the simultaneous influence of ROI location and FOV size on the difference between GV and HU
| ROI | FOV (cm) | Sample size | Mean GV (±SD) | Siga | Mean difference | 95% CI | Rank | |
|---|---|---|---|---|---|---|---|---|
| Lower limit | Upper limit | |||||||
| Anterior maxillary teeth | 8 × 8 | 10 | 1347.42 (±306.03) | 0.148 | 152.97 | -65.95 | 371.89 | 1 |
| 8 × 12 | 10 | 1580.00 (±210.43) | 0.000 | 385.55 | 235.02 | 536.08 | 2 | |
| 12 × 15 | 10 | 1659.78 (±185.78) | 0.000 | 465.33 | 332.43 | 598.23 | 3 | |
| Posterior maxillary teeth | 8 × 8 | 15 | 1033.13 (±299.21) | 0.000 | -368.46 | -534.16 | -202.477 | 3 |
| 8 × 12 | 15 | 1163.13 (±261.76) | 0.003 | -238.60 | -383.55 | -93.63 | 2 | |
| 12 × 15 | 15 | 1331.51 (±339.10) | 0.436 | -70.22 | -258.00 | 117.56 | 1 | |
| Anterior mandibular bone | 8 × 8 | 15 | 949.13 (±146.60) | 0.164 | -55.64 | -136.82 | 25.54 | 1 |
| 8 × 12 | 15 | 1157.61 (±142.74) | 0.001 | 152.84 | 73.79 | 231.88 | 2 | |
| 12 × 15 | 15 | 1366.45 (±206.40) | 0.000 | 361.68 | 247.38 | 475.97 | 3 | |
| Anterior mandibular teeth | 8 × 8 | 15 | 1026.57 (±290.97) | 0.737 | 25.77 | -135.36 | 186.91 | 1 |
| 8 × 12 | 15 | 1354.67 (±109.73) | 0.000 | 353.87 | 293.10 | 414.63 | 2 | |
| 12 × 15 | 15 | 1610.63 (±162.75) | 0.000 | 609.83 | 519.69 | 699.96 | 3 | |
| Posterior mandibular bone | 8 × 8 | 15 | 893.97 (±466.17) | 0.815 | 28.67 | -229.48 | 286.83 | 1 |
| 8 × 12 | 15 | 991.09 (±440.89) | 0.238 | 125.79 | -118.37 | 369.95 | 2 | |
| 12 × 15 | 15 | 1236.81 (±495.36) | 0.012 | 371.51 | 91.19 | 645.83 | 3 | |
CI = confidence interval; FOV = field of view, GV = gray value,ROI = region of interest, Sig = significance.
One-sample t-test due to normal distribution.
Table 6.
Comparing the simultaneous influence of ROI location and FOV size on the difference between GV and HU
| ROI | FOV (cm) | Sample size | Mean GV (±SD) | Siga | Difference from median | Rank |
|---|---|---|---|---|---|---|
| Anterior maxillary bone | 8 × 8 | 15 | 852.41 (±258.32) | 0.035 | 35.27 | 2 |
| 8 × 12 | 15 | 815.55 (±267.61) | 0.302 | 59.37 | 1 | |
| 12 × 15 | 15 | 998.59 (±286.03) | 0.000 | 214.97 | 3 | |
| Posterior maxillary bone | 8 × 8 | 15 | 648.54 (±401.03) | 0.302 | -190.17 | 2 |
| 8 × 12 | 15 | 867.03 (±412.57) | 1.00 | 23.13 | 1 | |
| 12 × 15 | 15 | 1058.47 (±456.74) | 0.035 | 259.63 | 3 | |
| Posterior mandibular teeth | 8 × 8 | 15 | 1781.32 (±267.46) | 0.007 | 159.3 | 1 |
| 8 × 12 | 15 | 1929.67 (±288.41) | 0.001 | 267.1 | 2 | |
| 12 × 15 | 15 | 2064.91 (±257.76) | 0.000 | 428.2 | 3 |
FOV = field of view, GV = gray value,ROI = region of interest, Sig = significance.
Two-sample t-test due to non-normal distribution.
Discussion
Although the existing evidence supports the strong correlation between GV and HU, clinicians are cautious about using these density scales interchangeably due to various factors that can introduce potential errors in assessing bone quality and making diagnoses. 8–19 Of these factors, the current study aimed to investigate FOV size and ROI location to discover their impact on GV measurement. Also, the difference between GV and HU (gold-standard) was calculated, which indicated the potential error in CBCT density assessment through GV. In summary, FOV size and ROI location significantly affected the magnitude of GV. The difference between GV and HU significantly differed within different ROIs. The anterior mandibular bone and posterior mandibular teeth ROIs resulted in the lowest and highest differences, respectively. Furthermore, the minimum error was recorded when 8 × 8 cm FOV was selected, while 12 × 15 cm FOV yielded the highest error.
In this study, 45 CBCT scans and 345 GV measurements were conducted, surpassing the sample sizes of previous studies conducted by Varshowsaz et al 23 and Caldas Mde et al. 24 While some studies used three to five piglet heads for their investigations, we opted to utilize three human skulls to obtain more applicable diagnostic results. 22,25,26 This decision was influenced by the fact that the physiological structure of human tissues can lead to degrees of scattered radiation, which may not be accounted for in studies using artificial phantoms. Hence, ignoring this factor can result in an overestimated reliability of the scanner. 11,17 Additionally, Emadi et al 27 measured the density of teeth, bones, and biomaterials using three imaging systems, including one spiral CT and two CBCT devices. It was found that NewTom VGi evo obtained superior results in discriminating dental materials, making it a powerful diagnostic tool in clinical practice. Thus, the mentioned imaging system was implemented in the current study. Notably, characteristics of the CBCT device might be influential on the resultant GVs. 12 As elucidated by Selvaraj et al, 28 each CBCT device possesses distinct calculations for density measurements. Therefore, experimenting with other scanning devices might not necessarily lead to the same outcomes.
Our findings indicated that, except for the posterior maxillary teeth ROI, the GVs obtained through CBCT were higher than the HUs obtained through MDCT in all ROIs. This observation was in line with the reports of Parsa et al, 17 indicating higher values in two CBCT systems. In a study conducted by Arisan et al, 29 the same finding was confirmed regarding the quantitative superiority of GVs compared to HUs. In contrast, Nackaerts et al 15 reported lower GVs in multiple CBCT systems compared to HUs. These discrepancies may be attributed to differences between the two imaging modalities in terms of cone beam geometry, the mathematical algorithm for image reconstruction, scattered radiation, noise, and artifacts. 12 In our study, the anterior mandibular bone exhibited the smallest difference between GV and HU, which can be explained by the superior visualization capabilities of CBCT in high-contrast structures and the smaller bone marrow spaces within this region, representing a D1 bone. 30–32
According to our findings, FOV size affected the GVs significantly. Increasing the FOV size resulted in higher GVs and a greater difference from HUs. This impact has been endorsed by various studies. Ibrahim et al 33 evaluated imaging under five FOVs, including 4 × 4 cm, 6 × 6 cm, 8 × 8 cm, 10 × 10 cm, and 10 × 5 cm. As a result, the size of FOV was identified as a determining factor in GV measurements. The mentioned influence was also expressed by Pauwels et al 12 when scanning a polymethyl methacrylate (PMMA) phantom using 13 CBCT and 1 MDCT devices. In another study by Katsumata et al, 34 the significant effect of FOV dimensions on GV was supported, noting that artifacts were intensified when more objects were present outside the FOV. Parsa et al 17 also reported an increase in GV with FOV enlargement in one of their investigated CBCT systems. Shokri et al 18 explained that a larger FOV could contribute to higher GVs. Moreover, as asserted by Campos et al, 35 smaller FOV resulted in lower GV due to the narrower diameter of X-ray beams, which merely irradiated the ROI and limited the exposure to surrounding tissues. This narrowing of the X-ray beams reduces the number of low-energy photons and enhances the penetration ability, potentially accounting for lower GVs. 19
Water has been commonly used as a phantom-filling material in various studies due to its ability to simulate soft tissues for radiographic purposes when used in thicknesses greater than 4 mm. 11,18,24 Nevertheless, there have been studies preferring ballistic gelatin as the material of choice for soft tissue stimulation. 25,26 However, using gelatin as the containing material has disadvantages such as more difficult removal from the inspected object.
While there have been improvements in the accuracy of GV for measuring bone densities and patterns, the reliability of this scale has been questioned by Pauwels et al 32 in a systematic review. Conversely, several studies have found a close relationship between GVs in CBCT and HUs obtained from MDCT. 8–10,36 It should be noted that HU may also exhibit a range of deviations when varying scanning protocols, different devices, or diverse energy settings within the same device are employed. 37–40 It has been highlighted that alterations in scanning protocols can result in discrepancies of up to 20% in HU values. Factors such as voltage, current, circulation time, slice thickness, and others can influence HU magnitude. Nevertheless, these discrepancies can be substantially mitigated using specific conversion tables and automated protocol recognition techniques. 39 Considering that the quality of cortical and cancellous bone directly affects the primary stability of implants and treatment success, CBCT appears to be a compelling choice for density measurements due to lower costs and radiation dose, as well as higher resolution and convenience. 32,36,41,42 The findings of our study provide oral and maxillofacial radiologists with valuable information regarding the regions that exhibit higher or lower differences between GV and HU as the gold-standard scale. This knowledge can help them select FOVs more accurately and position the patient appropriately to minimize discrepancies in GV measurements.
Regarding study limitations, a low number of dried human skulls were accessible for educational purposes, and finding skulls meeting our inclusion and exclusion criteria was challenging. Therefore, one of the included skulls lacked anterior maxillary teeth, hence this ROI could not be evaluated in this particular skull. Additionally, using water as the filling material for the phantoms may not perfectly mimic human soft tissue attenuation, and desiccated bone may differ from normal human tissues, which could introduce discrepancies in density measurements. Further studies should be conducted to evaluate different CBCT systems, DICOM viewers, FOVs, resolutions, voltages, and currents using materials that resemble soft tissue better or even on human cadavers.
Conclusion
The GVs significantly varied among investigated ROIs, with the lowest value recorded in the anterior maxillary bone ROI and the highest in the posterior mandibular teeth ROI. When GVs obtained from CBCT and gold-standard HUs were compared, anterior mandibular bone ROI showed the smallest difference. In contrast, posterior mandibular teeth ROI showed the largest difference, representing GV as a potentially erroneous scale in this region. Significant differences in GVs were also observed across different FOVs, with the 8 × 8 cm FOV resulting in the lowest GVs and the least difference from HUs, while the 12 × 15 cm FOV yielded the highest GVs with the greatest difference from HUs.
Footnotes
Contributors: AY, SS, SY: Acquisition, analysis, and interpretation of data, Drafting the manuscript, Final approval, Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. MGA: Analysis and interpretation of data, Revising the manuscript, Final approval, Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. YS: Conception and design, Revising the manuscript, Final approval, Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Contributor Information
Atiye Yadegari, Email: atiye.yad@gmail.com.
Yaser Safi, Email: yaser_safi@gmail.com, yaser_safi@yahoo.com.
Soheil Shahbazi, Email: soheil.shahbazi@gmail.com.
Sahar Yaghoutiazar, Email: saharyagotiazar@gmail.com.
Mitra Ghazizadeh Ahsaie, Email: mitraghazizadeh@gmail.com.
REFERENCES
- 1. Alamri HM, Sadrameli M, Alshalhoob MA, Sadrameli M, Alshehri MA. Applications of CBCT in dental practice: a review of the literature. Gen Dent 2012; 60: 390–400. [PubMed] [Google Scholar]
- 2. Lorenzoni DC, Bolognese AM, Garib DG, Guedes FR, Sant’anna EF. Cone-beam computed tomography and Radiographs in dentistry: aspects related to radiation dose. Int J Dent 2012; 2012: 813768. doi: 10.1155/2012/813768 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Roberts JA, Drage NA, Davies J, Thomas DW. Effective dose from cone beam CT examinations in dentistry. Br J Radiol 2009; 82: 35–40. doi: 10.1259/bjr/31419627 [DOI] [PubMed] [Google Scholar]
- 4. Petersen LB, Olsen KR, Christensen J, Wenzel A. Image and surgery-related costs comparing cone beam CT and panoramic imaging before removal of impacted Mandibular third molars. Dentomaxillofac Radiol 2014; 43(): 20140001. doi: 10.1259/dmfr.20140001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Benavides E, Rios HF, Ganz SD, An C-H, Resnik R, Reardon GT, et al. Use of cone beam computed tomography in implant dentistry: the International Congress of oral Implantologists consensus report. Implant Dent 2012; 21: 78–86. doi: 10.1097/ID.0b013e31824885b5 [DOI] [PubMed] [Google Scholar]
- 6. Safi Y, Aghdasi MM, Ezoddini-Ardakani F, Beiraghi S, Vasegh Z. Effect of metal artifacts on detection of vertical root fractures using two cone beam computed tomography systems. Iran Endod J 2015; 10: 193–98. doi: 10.7508/iej.2015.03.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Broder J, Preston R. Chapter 1 - Imaging the Head and Brain. In: Broder J, ed. Diagnostic Imaging for the Emergency Physician. Saint Louis: W.B. Saunders; 2011, pp. 1–45. [Google Scholar]
- 8. Razi T, Niknami M, Alavi Ghazani F. Relationship between Hounsfield unit in CT scan and gray scale in CBCT. J Dent Res Dent Clin Dent Prospects 2014; 8: 107–10. doi: 10.5681/joddd.2014.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kamaruddin N, Rajion ZA, Yusof A, Aziz ME. Relationship between Hounsfield unit in CT scan and gray scale in CBCT. AIP Conference Proceedings; Pinang, Malaysia. Vol. 1791; 2016. doi: 10.1063/1.4968860 [DOI] [Google Scholar]
- 10. Khojastepour L, Mohammadzadeh S, Jazayeri M, Omidi M. In vitro evaluation of the relationship between gray scales in Digital Intraoral Radiographs and Hounsfield units in CT scans. J Biomed Phys Eng 2017; 7: 289–98. [PMC free article] [PubMed] [Google Scholar]
- 11. Mah P, Reeves TE, McDavid WD. Deriving Hounsfield units using grey levels in cone beam computed tomography. Dentomaxillofac Radiol 2010; 39: 323–35. doi: 10.1259/dmfr/19603304 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Pauwels R, Nackaerts O, Bellaiche N, Stamatakis H, Tsiklakis K, Walker A, et al. Variability of dental cone beam CT grey values for density estimations. Br J Radiol 2013; 86: 1021: 20120135. doi: 10.1259/bjr.20120135 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Groh BA, Siewerdsen JH, Drake DG, Wong JW, Jaffray DA. A performance comparison of flat-panel Imager-based MV and kV cone-beam CT. Med Phys 2002; 29: 967–75. doi: 10.1118/1.1477234 [DOI] [PubMed] [Google Scholar]
- 14. Makins SR. Artifacts interfering with interpretation of cone beam computed tomography images. Dent Clin North Am 2014; 58: 485–95. doi: 10.1016/j.cden.2014.04.007 [DOI] [PubMed] [Google Scholar]
- 15. Nackaerts O, Maes F, Yan H, Couto Souza P, Pauwels R, Jacobs R. Analysis of intensity variability in Multislice and cone beam computed tomography. Clin Oral Implants Res 2011; 22: 873–79. doi: 10.1111/j.1600-0501.2010.02076.x [DOI] [PubMed] [Google Scholar]
- 16. Eguren M, Holguin A, Diaz K, Vidalon J, Linan C, Pacheco-Pereira C, et al. Can gray values be converted to Hounsfield units? A systematic review. Dentomaxillofac Radiol 2022; 51(): 20210140. doi: 10.1259/dmfr.20210140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Parsa A, Ibrahim N, Hassan B, van der Stelt P, Wismeijer D. Influence of object location in cone beam computed tomography (Newtom 5G and 3d Accuitomo 170) on gray value measurements at an implant site. Oral Radiol 2014; 30: 153–59. doi: 10.1007/s11282-013-0157-x [DOI] [Google Scholar]
- 18. Shokri A, Ramezani L, Bidgoli M, Akbarzadeh M, Ghazikhanlu-Sani K, Fallahi-Sichani H. 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. Imaging Sci Dent 2018; 48: 31–39. doi: 10.5624/isd.2018.48.1.31 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Rodrigues AF, Campos MJ da S, Chaoubah A, Fraga MR, Farinazzo Vitral RW. Use of gray values in CBCT and MSCT images for determination of density: influence of variation of FOV size. Implant Dent 2015; 24: 155–59. doi: 10.1097/ID.0000000000000179 [DOI] [PubMed] [Google Scholar]
- 20. Lagravère MO, Carey J, Ben-Zvi M, Packota GV, Major PW. Effect of object location on the density measurement and Hounsfield conversion in a Newtom 3G cone beam computed tomography unit. Dentomaxillofac Radiol 2008; 37: 305–8. doi: 10.1259/dmfr/65993482 [DOI] [PubMed] [Google Scholar]
- 21. Katsumata A, Hirukawa A, Okumura S, Naitoh M, Fujishita M, Ariji E, et al. Relationship between density variability and imaging volume size in cone-beam computerized Tomographic scanning of the Maxillofacial region: an in vitro study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009; 107: 420–25. doi: 10.1016/j.tripleo.2008.05.049 [DOI] [PubMed] [Google Scholar]
- 22. Santaella GM, Visconti M, Devito KL, Groppo FC, Haiter-Neto F, Asprino L. Evaluation of different soft tissue-Simulating materials in Pixel intensity values in cone beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol 2019; 127: e102–7. doi: 10.1016/j.oooo.2018.12.015 [DOI] [PubMed] [Google Scholar]
- 23. Varshowsaz M, Goorang S, Ehsani S, Azizi Z, Rahimian S. Comparison of tissue density in Hounsfield units in computed tomography and cone beam computed tomography. J Dent (Tehran) 2016; 13: 108–15. [PMC free article] [PubMed] [Google Scholar]
- 24. Caldas M de P, Ramos-Perez F de M, de Almeida SM, Haiter-Neto F. Comparative evaluation among different materials to replace soft tissue in oral Radiology studies. J Appl Oral Sci 2010; 18: 264–67: S1678-77572010000300012. doi: 10.1590/s1678-77572010000300012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Nascimento EHL, Fontenele RC, Lopes P de A, Santaella GM, Vasconcelos KF, de Freitas DQ, et al. Development of a model of soft tissue simulation using ballistic Gelatin for CBCT acquisitions related to Dentomaxillofacial Radiology research. Dentomaxillofac Radiol 2021; 50(): 20200191. doi: 10.1259/dmfr.20200191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Lopes PA, Santaella GM, Lima CAS, Vasconcelos K de F, Groppo FC. Evaluation of soft tissues Simulant materials in cone beam computed tomography. Dentomaxillofac Radiol 2019; 48(): 20180072. doi: 10.1259/dmfr.20180072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Emadi N, Safi Y, Akbarzadeh Bagheban A, Asgary S. Comparison of CT-number and gray scale value of different dental materials and hard tissues in CT and CBCT. Iran Endod J 2014; 9: 283–86. [PMC free article] [PubMed] [Google Scholar]
- 28. Selvaraj A, Jain RK, Nagi R, Balasubramaniam A. Correlation between gray values of cone-beam computed Tomograms and Hounsfield units of computed Tomograms: A systematic review and meta-analysis. Imaging Sci Dent 2022; 52: 133–40. doi: 10.5624/isd.20210274 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Arisan V, Karabuda ZC, Avsever H, Özdemir T. Conventional multi-slice computed tomography (CT) and cone-beam CT (CBCT) for computer-assisted implant placement. part I: relationship of radiographic gray density and implant stability. Clin Implant Dent Relat Res 2013; 15: 893–906. doi: 10.1111/j.1708-8208.2011.00436.x [DOI] [PubMed] [Google Scholar]
- 30. Pauwels R, Beinsberger J, Stamatakis H, Tsiklakis K, Walker A, Bosmans H, et al. Comparison of spatial and contrast resolution for cone-beam computed tomography scanners. Oral Surg Oral Med Oral Pathol Oral Radiol 2012; 114: 127–35. doi: 10.1016/j.oooo.2012.01.020 [DOI] [PubMed] [Google Scholar]
- 31. Misch CE. Contemporary implant dentistry. Implant Dentistry 1999; 8: 90. doi: 10.1097/00008505-199901000-00013 [DOI] [Google Scholar]
- 32. Pauwels R, Jacobs R, Singer SR, Mupparapu M. CBCT-based bone quality assessment: are Hounsfield units applicable Dentomaxillofacial Radiology 2015; 44: 20140238. doi: 10.1259/dmfr.20140238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ibrahim N, Parsa A, Hassan B, van der Stelt P, Aartman IHA, Wismeijer D. The effect of Scan parameters on cone beam CT trabecular bone Microstructural measurements of human mandible. Dentomaxillofac Radiol 2013; 42(): 20130206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Katsumata A, Hirukawa A, Okumura S, Naitoh M, Fujishita M, Ariji E, et al. Effects of image artifacts on gray-value density in limited-volume cone-beam computerized tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007; 104: 829–36. doi: 10.1016/j.tripleo.2006.12.005 [DOI] [PubMed] [Google Scholar]
- 35. Campos MJ da S. Bone mineral density in cone beam computed tomography: only a few shades of gray. WJR 2014; 6: 607. doi: 10.4329/wjr.v6.i8.607 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Razi T, Emamverdizadeh P, Nilavar N, Razi S. Comparison of the Hounsfield unit in CT scan with the gray level in cone-beam CT. J Dent Res Dent Clin Dent Prospects 2019; 13: 177–82. doi: 10.15171/joddd.2019.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Lamba R, McGahan JP, Corwin MT, Li C-S, Tran T, Seibert JA, et al. CT Hounsfield numbers of soft tissues on Unenhanced abdominal CT scans: variability between two different manufacturers' MDCT scanners. AJR Am J Roentgenol 2014; 203: 1013–20. doi: 10.2214/AJR.12.10037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Jeong D-K, Lee S-S, Kim J-E, Huh K-H, Yi W-J, Heo M-S, et al. Effects of energy level, reconstruction kernel, and tube rotation time on Hounsfield units of Hydroxyapatite in virtual Monochromatic images obtained with dual-energy CT. Imaging Sci Dent 2019; 49: 273–79. doi: 10.5624/isd.2019.49.4.273 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Zurl B, Tiefling R, Winkler P, Kindl P, Kapp KS. Hounsfield units variations: impact on CT-density based conversion tables and their effects on dose distribution. Strahlenther Onkol 2014; 190: 88–93. doi: 10.1007/s00066-013-0464-5 [DOI] [PubMed] [Google Scholar]
- 40. Sorooshfard E, Tahmasbi M, Chegeni N, Tahmasebi Birgani MJ. Evaluating the effects of variation in CT scanning parameters on the image quality and Hounsfield units for optimization of dose in radiotherapy treatment planning: A semi-anthropomorphic Thorax phantom study. J Cancer Res Ther 2023; 19: 426–34. doi: 10.4103/jcrt.jcrt_260_21 [DOI] [PubMed] [Google Scholar]
- 41. Greenberg AM. Advanced dental implant placement techniques. J Istanb Univ Fac Dent 2017; 51: S76–89. doi: 10.17096/jiufd.17594 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Ronkainen A-P, Al-Gburi A, Liimatainen T, Matikka H. A dose–neutral image quality comparison of different CBCT and CT systems using Paranasal sinus imaging protocols and phantoms. Eur Arch Otorhinolaryngol 2022; 279: 4407–14. doi: 10.1007/s00405-022-07271-4 [DOI] [PMC free article] [PubMed] [Google Scholar]