Lab-based X-ray tomography of a cochlear implant using energy discriminating detectors for metal artefact reduction

X-ray computed tomography (XCT) is an important clinical diagnostic tool which is also used in a range of biological imaging applications in research. The increasing prevalence of metallic implants in medical and dental radiography and tomography has driven the demand for new approaches to solving the issue of metal artefacts in XCT. Metal artefacts occur when a highly absorbing material is imaged which is in boundary contact with one or more weakly absorbing components, such as soft-tissue. The resulting 'streaking' in the reconstructed images creates significant challenges for X-ray analysis due to the non-linear dependence on the absorption properties of the sample. In this paper we introduce a new approach to removing metal artefacts which exploits the capabilities of the recently available, photon-counting PiXirad detector. Our approach works for standard lab-based polychromatic X-ray tubes and does not rely on any post-processing of the data. The method is demonstrated using both simulated data from a test phantom and experimental data collected from a cochlear implant. The results show that by combining the individual images, which are simultaneously generated for each different energy threshold, artefact-free segmentation of the implant from the surrounding biological tissue is achieved.