To achieve accurate dose calculations in radiation therapy the electron density of patient tissues must be known. This information is ordinarily gained from a computed tomography (CT) image that has been calibrated to allow relative electron density (RED) to be determined from CT number. When high density objects such as metallic prostheses are involved, direct use of the CT data can become problematic due to the artefacts introduced by high attenuation of the beam. This requires manual correction of the density values, however the properties of the implanted prosthetic are not always known. A method is
introduced where the RED of such an object can be determined using the treatment beam of a linear accelerator with an electronic portal imaging device. The technique was tested using a metallic hip replacement that was placed within a container of water. Compared to the theoretical RED of 6.8 for cobalt–chromium alloy, these measurements calculated a value of 6.4 ± 0.7. This would allow the distinction of an implant as Co–Cr or steel, which have similar RED, or titanium, which is much less dense with an RED of 3.7.