How good are radiographs constructed from low dose CBCT?
The introduction of CBCT imaging has the potential to transform orthodontic imaging. However, concerns remain about its comparatively high radiation dose compared to traditional two-dimensional imaging. Nevertheless, in recent years, advancements in CBCT technology have aimed to minimise the radiation exposure for children. This may allow us to use panoramic and cephalometric images extracted from CBCT images.
However, there has been limited research into the image quality of potential reduced-dose radiographs. This paper provides us with valuable information on this subject.
A multinational predominantly based in the Netherlands did this study. The American Journal of Orthodontics published the paper.
Rogier H. van Bunningen, Pieter U. Dijkstra, Arjan J. A. Dieters, Paul van der Stelt, Wicher J. van der Meer, and Anne Marie Kuijpers-Jagtman.
AJO-DDO advance access. https://doi.org/10.1016/j.ajodo.2025.09.025
This is an Open Access paper, so we can all read it.
I have an interest to declare as I have known Anne Marie Kuijpers-Jagtman for many years as an academic colleague.
What did they ask?
They did this study to.
“ Evaluate the image quality of Standard dose Panoral (PAN), extracted PAN and reduced dose PAN. For orthodontic applications”.
What did they do?
They carried out a rather complex study, and I hope I have explained it adequately. Please refer to the paper for further clarification.
They did the study in the following stages.
- They collected routine standard dose CBCT scans and standard dose PANS (sd-PAN) from the existing database of their department. This resulted in 43 sets of records.
- They then extracted a further PAN (extr-PAN) from the CBCT.
- Then they manipulated the (extr-PAN) using a precise filtering technique to construct a reduced dose PAN. This was called the sim rd-PAN.
- They then asked three orthodontists who each had greater than 20 years of experience in interpreting orthodontic radiographs to assess the three types of images of the 43 patients.
They presented the following statements to the three orthodontists.
The image is:
- Technically acceptable and usable
- Is suitable for assessing the course of the mandibular canal.
- Is suitable for assessing the condyles
- Is suitable for assessing the floor of the maxillary sinus.
- An orthodontic treatment plan can be based on this plan
The assessors indicated whether or not they agreed with the statement with a yes/no response..
They also presented them with three subjective statements. These were
- The germ of the third molar is present.
- The root of an identified tooth is fully developed
- The roots of the teeth are parallel.
- Fixed appliances in the upper or lower jaw are present.
They compared these latter statements with a gold standard that they developed.
They performed relevant statistical analysis to evaluate agreement among the examiners and different levels of concordance with the statements.
What did they find?
They provided extensive details on the examiners’ agreement with the statements. I cannot go into these now, but generally, they reported that only the sd-PAN met the threshold for technical adequacy and suitability for orthodontic treatment planning.
However, the extr-PAN and the sim rd-PAN met the threshold for acceptable quality for statements 6,7 and 9.
Their overall conclusion was:
“Although the extracted PAN and simulated reduced-dose PAN, showed limited diagnostic image quality for orthodontic diagnosis. The differences in the percentage of agreement with a gold standard for the various images were small. Therefore, given the low radiation exposure and added value of 3-D information, the ultra low-dose CBCT, the extracted PAN, and the reduced dose PAN might be considered promising alternatives for orthodontic diagnosis. Further research is needed to determine their diagnostic value”.
I normally shorten the conclusions; however, I felt it was important to include their entire conclusion here to aid understanding.
What did I think?
I have been involved in several studies that examine the effectiveness of imaging and image quality on orthodontic treatment decisions. These have been quite complex to conduct. I would like to point out that the authors have carried out this study very well, and I hope I have managed to interpret their methodology and their findings correctly.
They employed a clear and practical methodology that led to some interesting findings. I found it particularly important to note that the extracted panoramic radiographs had lower image quality than the standard dose radiographs. This was not surprising; what is interesting is that the panel of orthodontists considered these images suitable for developing orthodontic treatment plans.
This finding is quite significant and may mirror clinical practice and experience. I am sure that we all possess expertise in interpreting views that are less than “ideal”.
There clearly needs to be more research carried out in this area. This particularly requires studies that examine orthodontic treatment decisions based on the various types of images that we can use.
As a result, this study is of great value and guides future research that is essential before we adopt CBCT imaging as our standard first-line radiographic examination. So let’s still be careful with that radiation!
Emeritus Professor of Orthodontics, University of Manchester, UK.