Login
- Dokumenttyp:
- Zeitschriftenaufsatz
- Autor(en):
- Esteban, J.; Grimm, M.; Unberath, M.; Zahnd, G.; Navab, N.
- Titel:
- Towards fully automatic X-ray to CT registration
- Abstract:
- The main challenge preventing a fully-automatic X-ray to CT registration is an initialization scheme that brings the X-ray pose within the capture range of existing intensity-based registration methods. By providing such an automatic initialization, the present study introduces the first end-to-end fully-automatic registration framework. A network is first trained once on artificial X-rays to extract 2D landmarks resulting from the projection of CT-labels. A patient-specific refinement scheme is then carried out: candidate points detected from a new set of artificial X-rays are back-projected onto the patient CT and merged into a refined meaningful set of landmarks used for network re-training. This network-landmarks combination is finally exploited for intraoperative pose-initialization with a runtime of 102ms. Evaluated on 6 pelvis anatomies (486 images in total), the mean Target Registration Error was 15.0+-7.3mm. When used to initialize the BOBYQA optimizer with normalized cross-correlation, the average (+-STD) projection distance was 3.4 +- 2.3mm, and the registration success rate (projection distance <2.5% of the detector width) greater than 97%.
- Stichworte:
- X-ray to CT Registration,Projective Geometry,Neural Network,Patient-Specific Training
- Jahr:
- 2019
BibTeX