Registration of pre- and intra-interventional data is the key technology for image-guided radiation therapy, minimally invasive surgery, endoscopy, and interventional radiology. A survey of those 3D/2D data registration methods is presented which utilize 3D CT or MR images as the pre-interventional data and 2D X-ray projection images as the intra-interventional data. The 3D/2D registration methods are reviewed with respect to image modality, image dimensionality, registration basis, geometric transformation, user interaction, optimization procedure, and object of registration.
COBISS.SI-ID: 7682900
A novel 3D/2D registration method is proposed that is based on robustly matching 3D pre-interventional image gradients and coarsely reconstructed 3D gradients from 2D gradients of the intra-interventional images. The proposed method is evaluated using the publicly available standardized evaluation methodology, consisting of 3D rotational X-ray, computed tomography, magnetic resonance (MR), and 2D X-ray images of two spine segments, and standardized evaluation criteria. The proposed method is more accurate and robust that the other state-of-the art methods.
COBISS.SI-ID: 6611540
A new standardized evaluation methodology, consisting of an image database, protocol and metrics, for objective evaluation of 3D/2D registration methods was introduced. The database comprised CT images of the pelvis and spine of a male and female from the Visible Human Project, and 6 pairs of 2D fluoroscopic x-ray images rendered from the CT. A 3-gauge needle was simulated in the 2D images. The proposed methodology was used to evaluate two gradient-based 3D/2D registration methods according to accuracy, robustness, and complexity.
COBISS.SI-ID: 8096084
A gold standard data set was produced using a fresh cadaver pig head with attached fiducial markers. The head was imaged with imaging modalities common in diagnostic imaging or radiotherapy, which include 64-slice CT, MRI using T1, T2, and proton density sequences, and cone beam CT imaging data. Radiographic data were acquired using kilovoltage and megavoltage imaging techniques. The image information reflects both anatomy and reliable fiducial marker information and improves over existing data sets by the level of anatomical detail, image data quality, and soft-tissue content.
COBISS.SI-ID: 8246868
The performance of an intensity-based 2D/3D registration method with four merit functions, namely, cross correlation, rank correlation, correlation ratio, and mutual information (MI), and two gradient-based algorithms, the backprojection and the reconstruction registration method were tested using the new phantom data set. The results indicate that gradient-based methods are more accurate than intensity-based methods, especially when soft tissue deformation is present.
COBISS.SI-ID: 8247124