Related Experiment Video
Updated: May 21, 2026

Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Auto-masked 2D/3D image registration and its validation with clinical cone-beam computed tomography
P Steininger1, M Neuner, H Weichenberger
1Institute for Research and Development on Advanced Radiation Technologies (radART), Paracelsus Medical University, Muellner Hauptrasse 48, 5020 Salzburg, Austria. philipp.steininger@pmu.ac.at
A novel 2D/3D image-guided alignment method accurately registers patient anatomy using X-rays and CT scans, offering a viable alternative to cone-beam CT for radiotherapy setup. This technique ensures precise patient positioning for improved treatment delivery.
Area of Science:
- Medical Physics
- Radiotherapy Technology
- Image Processing
Background:
- Image-guided radiotherapy (IGRT) is crucial for minimizing setup errors between planned and actual patient positioning.
- Accurate patient alignment is essential for effective radiation dose delivery and minimizing off-target exposure.
- Current IGRT methods, such as cone-beam CT (CBCT), involve volumetric data acquisition, which can increase treatment time and radiation dose.
Purpose of the Study:
- To develop and validate a 2D/3D image registration approach for precise patient alignment in radiotherapy.
- To evaluate the accuracy and reliability of the 2D/3D method compared to standard 3D/3D CBCT-to-CT alignment.
- To assess the feasibility of using X-ray images and CT data for rigid alignment of pelvic anatomy.
Main Methods:
- A novel 2D/3D registration algorithm was developed, rigidly aligning computed tomography (CT) data with orthogonal X-ray images.
- The algorithm maximizes pixel intensity agreement between X-rays and CT-derived radiographs, using rank correlation as the intensity metric.
- Regions of interest were selected using 3D segmentations from pre-planning CT data, and validation was performed on 80 pelvic CBCT datasets.
Main Results:
- The 2D/3D registration achieved high accuracy, with root mean square errors <0.18° for rotations and <0.73 mm for translations.
- A mean target registration error of <2 mm in the lesser pelvis was achieved in 94.1% of over 8400 experiments.
- The 2D/3D method demonstrated comparable accuracy to standard 3D/3D CBCT-to-CT alignments for bony anatomy.
Conclusions:
- The developed 2D/3D registration method provides a viable and accurate alternative to CBCT-based alignment for pelvic radiotherapy.
- This approach is suitable when rigid alignment of bony anatomy is sufficient and volumetric data acquisition is not required.
- The method's accuracy is well within acceptable limits for many radiotherapy prescriptions, offering potential benefits in terms of speed and radiation dose.
Related Concept Videos
Imaging Studies III: Computed Tomography
Computed Tomography
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...

