Dosimetry analysis of panoramic‐imaging devices in different‐sized phantoms

Abstract The aim of this study is to measure the radiographic dose in adult, adolescent, and child head‐sized PMMA phantoms for three panoramic‐imaging devices: the panoramic mode on two CBCT machines (Carestream 9300 and i‐CAT NG) and the Planmeca ProMax 2D. A SEDENTEXCT dose index adult phantom and custom‐built adolescent and pediatric PMMA dosimetry phantoms were used. Panoramic radiographs were performed using a Planmeca ProMax 2D and the panoramic mode on a Carestream 9300 CBCT and an i‐CAT NG using the protocols used clinically. Point dose measurements were performed at the center, around the periphery and on the surface of each phantom using a thimble ionization chamber. Five repeat measurements were taken at each location. For each machine, single‐factor ANOVA was conducted to determine dose differences between protocols in each phantom, as well as determine the differences in absorbed dose when the same protocol was used for different‐sized phantoms. For any individual phantom, using protocols with lower kVp, mA, or acquisition times resulted in statistically significant dose savings, as expected. When the same protocol was used for different‐sized phantoms, the smaller phantom had a higher radiation dose due to less attenuation of x‐rays by the smaller phantom and differences in the positioning of the ion chamber relative to the focal trough. The panoramic‐mode on the CBCT machines produce images suitable for clinical use with similar dose levels to the stand‐alone panoramic device. Significant dose savings may result by selecting age‐ and size‐ appropriate protocols for pediatric patients, but a wider range of protocols for children and adolescents may be beneficial.

image obtained from a CBCT dataset and reformatting within the software, which may remove some of the superposition of structures that are expected in the panoramic image. Software reformatting provides a series of useful images from a single acquisition (3D volumes, 2D slices through anatomy of interest, 2D panoramic view, etc.) but clinicians must remember that the dose for the 3D acquisition is higher than for panoramic imaging 2 and this technique should only be done if the 3D images are required. However, some CBCT machines also offer a panoramic acquisition to obtain a true 2D image that is advertised as comparable with stand-alone machines, with some manufacturers' boasting a reduction in dose over standalone panoramic-imaging systems. To date, no dosimetry studies have been reported in the scientific literature to compare the dose received by the patient or the dose distribution in combination units compared with single-function panoramic-imaging machines.
Panoramic machines often use pre-set imaging protocols with various exposure parameters (kVp, mA, acquisition time) that determine the radiation dose to a patient. Helmrot et al. have suggested using the dose area product (DAP) as a standardized dose metric for all dental radiography 3 due to the convenience of measurements and that the DAP is measured independent of the patient and can therefore be specified by the manufacturer. The DAP has been used to establish radiation output reference levels in Greece 4 and Germany, 5 although the reference levels are highly dependent on the measurements used to find the 75% dose level, but are not for defining absorbed or effective doses. Although the DAP values are obtained at the tube port in an empty field, and therefore not including the beam dispersion or scattering effects within the patient, they have been used to estimate the effective doses for patients using published conversion factors. 3,5 However, the results are highly variable depending on the measurement techniques and whether the salivary glands were included in the effective dose calculations; 5 the tissueweighting factors were updated in ICRP 2007, which assigned a weighting factor to the salivary glands instead of including them in the remainder tissues. 6 Roberts et al. have also shown that effective doses calculated using older tissue-weighting factors (from 1990) for dental imaging are roughly half that using the factors published in 2007, six primarily due to the inclusion of the salivary glands in the calculation, 7 which limits the applicability of effective dose measurements. We propose using the absorbed dose measurement within a head-sized PMMA phantom, as it represents the energy absorbed within the phantoms, including dose from both the primary and scattered radiation, and can be used to estimate other metrics if desired.
Although dental imaging contributes less than 0.1% of the radiation dose the global population receives, radiation risk should always be considered when conducting panoramic radiography. 2 The radiation risk is three times greater in patients that are less than 10 yr old compared to those that are above 30 yr. 8 The increased radiosensitivity of tissues in children, along with their longer anticipated life span post-exposure, increases their risk of developing cancer over their lifetime. 9 The radiosensitive nature of pediatric patients validates the need to carefully monitor the radiation exposure to these patients in particular. There are very few studies examining the radiographic dose on pediatric patients from panoramic radiography, and none when using the panoramic-mode on a CBCT unit. A study conducted by Hayakawa et al. examined the doses in a dry-skull phantom representing a 5-6 yr old child for two single-function panoramic machines. 10 Comparing adult and child imaging protocols for the phantom, Hayakawa et al. concluded that pediatric exposure settings reduce dose irrespective of machine.
Choi et al. have developed two pediatric head-sized PMMA phantoms, representing a child aged 5 yr and an adolescent aged 12 yr, and measured the absorbed dose in various locations in dental CBCT, 11 leading to the same conclusions that pediatric exposure settings could dramatically decrease patient doses.
The aim of this study was to measure and compare the absorbed dose in adult, adolescent, and child head-sized PMMA phantoms for three panoramic-imaging devices: the panoramic modes on the Carestream 9300 CBCT and i-CAT Next Generation CBCT, and the Planmeca ProMax 2D panoramic machine. The study also aims to establish the importance of selecting patient-appropriate protocols, particularly in pediatric patients. Agency, Genome Sciences Center, Vancouver, Canada). The adolescent phantom (135 mm diameter 9 150 mm height) was designed to represent a 12-year-old child beginning orthodontic treatment, whereas the child phantom (100 mm diameter 9 150 mm height) was designed to represent a 5-year-old child. 11 The dimensions of the custom-built phantoms were obtained from measuring anatomic reference points in the dental CBCT images of pediatric patients.

2.A | Phantoms
A panoramic radiograph of a Pan DXTTR (Rinn Corporation, Elgin IL, USA) was taken by each machine to assess their images. The DXTTR phantom is an anthropomorphic phantom comprised of a natural bone skull embedded in resin. The head has detailed facial features to enable positioning using anatomical landmarks and is mounted on a tripod that can articulate to enable angling the head to align with the laser positioning guides.

2.B | Imaging systems
Panoramic radiographs were performed using the panoramic-mode A PMMA plate was mounted onto a tripod upon which the PMMA phantoms were situated. Positioning lasers were used to place each PMMA phantom within the field of view (FOV). Specifically, the front peripheral hole within the phantom was positioned in the imaging focal trough and the phantom was centered vertically within the field of view. Phantoms were exposed to acquisition protocols deemed clinically appropriate with regard to phantom dimensions. All protocols used in the study are pre-set within their respective machines except the small child protocol (ProMax 2D) for which the exposure parameters (kVp, mA, and acquisition time) were manually adjusted to match the posted clinically accepted technique chart. Table 1 shows the exposure parameters used for each protocol for the panoramic-mode on the CBCT machines and the standalone machine, respectively.
The Pan DXTTR phantom was situated within each device using positioning lasers to ensure that the mandible of the phantom was

2.C | Dosimetry measurements
A thimble ion chamber (10 9 6-0.6-CT; Radcal Corporation, Monrovia, CA), with an active volume of 0.6 cm 3 , was placed in either   Sidak's multiple comparisons test was conducted using the normalized absorbed dose value (lGy/mAs) to determine statistically significant differences between the Carestream 9300, i-CAT and ProMax 2D for equivalent protocols used for each phantom.

| RESULTS
The absorbed dose in adult-, adolescent-, and child-sized phantoms was measured for the panoramic-mode of the Carestream 9300 (Table 2), the panoramic-mode of the i-CAT NG (  Fig. 3, the adolescent phantom in Fig. 4 (Fig. 4).
Panoramic radiographs of the Pan DXTTR phantom using the average adult settings on the Promax 2D and Carestream 9300, and the large setting on the i-CAT NG are shown in Fig. 6. The images from the Promax 2D (Fig 6c) and the CS9300 (Fig. 6a)  when obtaining a panoramic image; for panoramic machines, the center of rotation changes during the acquisition, as opposed to a CBCT acquisition, which has a fixed center of rotation. We believe that the dose distribution indicates that the center of rotation changes throughout the scan when the CBCT machines operate in the panoramic acquisition mode. Clinicians must be also aware that the rotation of the x-ray tube behind the patient results in a non-uniform radiation dose distribution within a patient. The radiation doses were the lowest at the front of the phantom for peripheral and surface measurements in each machine due to the trajectory of the x-ray tube as expected. In the adult phantom, the highest doses were generally measured at the lateral peripheries for all devices. In larger phantoms, the sides are closer to the x-ray tube during its rotation increasing the radiation per unit area. The temple support is also in closer proximity to the sides of the adult phantom, facilitating radiation scatter at those peripheries.
The highest dose was measured at the center in both adolescent and child phantoms for the Carestream 9300 and ProMax 2D. X-rays pass through the center of these smaller diameter phantoms throughout the x-ray tube's trajectory, increasing the measured dose. X-rays must also travel a greater distance before reaching the adolescent and child   attenuation of x-rays through the PMMA also increases the radiation dose measured at the center and front of the smaller phantoms.
Clearly, children may be exposed to unnecessary radiation dose if exposure parameters are not adjusted appropriately. Minimizing radiation exposure to children is of primary concern due to the The bony structures will alter the absorption and scatter properties of the phantom, giving a more realistic dose distribution. Furthermore, having tissue equivalent material will also enable estimates of the radiation risk using ICRP weighting factors.

| CONCLUSION
The panoramic-mode on the CBCT machines studied produced diagnostic quality images with comparable radiographic dose to a stand-alone panoramic-imaging device. Smaller phantoms receive more radiation when imaging protocols are identical for each device.
The study demonstrates that pediatric protocols reduce the radiographic dose to children, but the combination units had a limited number of protocols available. All panoramic-imaging devices, both stand-alone and combination units, will benefit from including a larger range of pre-set options representing the pediatric and adolescent populations.  is produced by a dedicated panoramic radiographic unit. It is clear that the almost wholly dentate human skull (which includes all four impacted third molar teeth) imbedded in this DXXTR phantom has a missing lower incisor. Although (c) displays the full length of all teeth including the incisors, indicating that they (and those of (a)) are within the anterior focal trough for these units, whereas those in (b) display only the crowns, but not the apices which are outside the anterior focal trough. (b) displays more obviously the superimposition of the vertebral column and secondary images of the contralateral mandible. These differences reflect the standard positioning of the phantom as a patient in each of the three units during the exposure.

CONF LICT OF I NTEREST
The authors have no conflicts of interest to disclose.