Logo of the Physikalisch-Technische Bundesanstalt

Current state of the departmental research project for the determination of conversion factors for dose quantities in digital volume tomography


Since July 2019, Working Group 6.25 has been dealing with a project to develop a concept for the conversion of dose‑relevant parameters in computed tomography (CT) and cone‑beam computed tomography (CBCT). The dosimetry of these two methods is based on conceptually different quantities, which cannot be trivially standardized. These quantities are the dose length product (DLP) for CT and the dose area product (DAP) for CBCT.

The cone‑beam computed tomography (CBCT) procedure generates a three‑dimensional image of a patient with the aid of a rotating combination of an X‑ray tube and an image amplifier (or array detector). CBCT is a further development of conventional computed tomography, with the considerable difference that a broad cone beam is used. In this way, a wide image of the patient can be scanned in just one single full or partial rotation. Due to this, it is no longer necessary to move the table – which is required for conventional CT – to be able to scan the patient in the lateral direction. The scanners can therefore be manufactured at lower cost and can also be moved, which opens up new scanning possibilities. Today, cone‑beam CT scanners are being used for post‑operative checks by taking images as early as in the operating room: for 3D imaging during interventions (e.g. in the field of angiography); for breast cancer screening in dedicated breast CT scanners; and for dental images.

For quality assurance and to estimate the effective dose, a different parameter is used for CBCT than for CT. In CT dosimetry, the dose is determined by measuring the volume computed tomography dose index (CTDIvol) and the dose length product (DLP). Both of these are quantities implemented by means of a pencil ionization chamber in a cylindrical Plexiglas phantom, the so‑called CTDI phantom. However, in CBCT, the dose area product (DAP) is used as a parameter, which is the product of a dose and an irradiated surface.

The quantities of CTDIvol and DAP cannot be trivially converted from one system to another, as they differ as regards their concept. However, a conversion is desirable for the investigations which can be carried out with the aid of CT and CBCT. Among other things, this is important to determine the diagnostic reference levels (DRLs). DRLs are dose values for typical X‑ray examinations and are derived by the German Federal Office for Radiation Protection (Bundesamt für Strahlenschutz, BfS) from irradiation data provided by medical bodies at regular intervals. The DRLs help to detect and prevent values which, over a longer period of time, unjustifiably exceed a mean dose that has been laid down by DRLs.

Since July 2019, Working Group 6.25 has been cooperating with the Klinikum Braunschweig (Braunschweig’s municipal hospital) in a departmental research project of the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU), accompanied by the German Federal Office for Radiation Protection (BfS), to work on the comparability of these dose quantities. For this purpose, investigations are to be observed for which DRLs already exist and which are carried out in both methods. The results shall then be included in the publication of the DRLs.

The conversion is determined in a combined metrological and simulation-based procedure. For this purpose, those source parameters of the CT or CBCT X-ray sources which are not predefined in the device specifications, are determined by means of a mobile setup developed at PTB. With the aid of these source parameters, simulations can be carried out with the ImpactMC software to determine the effective dose on a standard patient. The conversion factors are to be determined by comparing the effective doses and the simultaneously calculated values for CTDIvol and DAP. To validate the procedure, measurements with MOSFET sensors are carried out simultaneously on an anthropomorphic phantom. The MOSFET sensors used have a diameter of only 3 mm and thus an excellent spatial resolution.

In one of the first steps, the respective literature was reviewed in order to find out the current state of research and the methodological approaches used so far. The search was carried out on the basis of the PRISMA method which provides specifications on how to implement the search and present the results. It was realized on the five search engines of PubMed, Web of Science, Scopus, Science Direct and EuropePMC. Between 1 January 2000 and 31 December 2019, 116 publications were found which deal with this problem, eight publications of which report on the conversion of dose values for CBCT or CT devices. However, the comparability of these studies is very limited, as different scan protocols and measurement methods were used and many publications require an improved way of reporting. A detailed study on conversion factors has so far not been carried out.

Parallel to that, the relevant devices and investigation methods were identified. The CBCT devices can be divided into three categories: C‑shaped scanners (so‑called C bends) which are firmly fixed to the ceiling or on the ground and can be mainly used for angiography; smaller, mobile C bends which are primarily used in intraoperative interventions; as well as more exotic devices. Eight representative devices were selected from these. They were compared with two CT scanners – one at the Working Group and one at the Klinikum Braunschweig. For this research project, investigations of the cranium after a stroke, of the paranasal sinuses during a sinusitis infection and of the cervical and lumber spine during a neurosurgical intervention have been regarded as relevant for the project.

Since January 2020, the metrological campaign has been underway, with measurements at the Klinikum Braunschweig and at the research CT scanner of the Working Group. The measurements were originally planned to run till spring 2021, while delays due to access restrictions at the Klinikum Braunschweig have not been ruled out because of the current Covid‑19 situation.

An initial first segmented data set of CT scans of an average patient is currently being set up. As soon as the data set has been completed, simulations of the effective dose will be created on the basis of the measured source parameters. The MOSFET sensors required for validation are characterized at the same time at the Working Group's X‑ray facilities. Preliminary results show a very strong dependence on the mean energy of the radiation, which will make a calibration of the sensors in the field of the CT scanner necessary. The dependence on the rotation angle, on the dose rate and on the accumulated dose are comparatively small.

The German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU), represented by the German Federal Office for Radiation Protection (BfS), is supporting the departmental research project. The project number is 3619S42462.

Your contact at PTB:

Opens local program for sending emailDr. Steffen Ketelhut, Department 6.2, Working Group 6.25

Opens local program for sending emailL. Büermann, Department 6.2, Working Group 6.25