Enhancement of patient-motion-disturbed (dental) Cone Beam Computed Tomography (CBCT) data.
Epidemiology and Medical Biometry/Statistics
Final Report Abstract
The project deals with motion correction in dental cone beam computed tomography (CBCT). More specifically, a solution was developed for the case of articulated mandibular movements, i.e. separate mandibular and cranial movements. In this project two different methods were developed, one based on segmentation of the lower jaw from the rest of the skull and separate but rigid motion correction. A second method is based on inverse neural rendering for reconstruction and a simultaneous optimisation of a deformation field, which in theory can represent any type of movement. The first method was successfully applied to dental data, while the second method was even able to capture and visualise respiratory movements in thoracic CBCTs. A quantitative comparison with other methods for the reconstruction of thoracic CBCTs showed that the developed method produces results at or above state-of-the-art with regard to various quality metrics.
Link to the final report
https://openscience.ub.uni-mainz.de/handle/20.500.12030/11725
Publications
-
Motion correction for separate mandibular and cranial movements in cone beam CT reconstructions. Medical Physics, 50(6), 3511-3525.
Birklein, Lukas; Niebler, Stefan; Schömer, Elmar; Brylka, Robert; Schwanecke, Ulrich & Schulze, Ralf
-
Neural Deformable Cone Beam CT. VCBM 2023
Birklein L., Schömer E., Brylka R., Schwanecke U. & Schulze R.
