To align patients in radiation devices in six degrees of freedom (DoF),image-guided approaches perform the task of correction computationfor the patient position. Digital radiography (DR) images are comparedto projections of a CT series to estimate misalignments. A problemis that digital reconstructed radiographs (DRRs) have to be createdfrom the CT to be registered with the DRs. Depending on the X-raytube energy, detector sensitivity and body part involved, DRRs andDRs may look very different and often cannot be registered. We presenta method that reconstructs multi-spectral DRRs for different X-raysettings, which can be registered to real X-ray images. As shortrendering times are crucial, multiple spectra of a DRR are generatedin one ray-tracing process. We register our multi-spectral DRR withthe DR and add a further DoF to find a best match not only for thetranslations and in-plane rotation, but also the best fitting spectralplanes. The results are used to identify patient misalignments andshow that higher reliability can be achieved compared to conventionalapproaches. Misalignments can be identified even if ineligible X-raysettings have been used. As our approach allows application of lowerX-ray energies for DR creation, an additional benefit is the reductionof the delivered dose.
«
To align patients in radiation devices in six degrees of freedom (DoF),image-guided approaches perform the task of correction computationfor the patient position. Digital radiography (DR) images are comparedto projections of a CT series to estimate misalignments. A problemis that digital reconstructed radiographs (DRRs) have to be createdfrom the CT to be registered with the DRs. Depending on the X-raytube energy, detector sensitivity and body part involved, DRRs andDRs may look very different a...
»