Data for: Accelerating B-spline Interpolation on GPUs: Application to Medical Image Registration

Published: 10 April 2020| Version 1 | DOI: 10.17632/kj3xcd776k.1
, Orestis Zachariadis


The dataset contains 5 pairs of images (for a total of 10 images). The images have been preprocessed, segmented and masked. Affine transformation has been applied across pre- and intra-operative pairs, i.e they are ready for non-rigid registration. There are 3 pairs of images of a liver phantom based on patient specific data. Deformation simulating pneumoperitoneum has been applied to these images. ARTORG centre and Cascination produced the phantom. They are x-ray reconstructions (DynaCT). Images 6 - 9 are image pairs of pre- and intra- operative porcine liver MRI scans conducted at Oslo University Hospital - Rikshospitalet, Norway. Intra-operative images have pneumoperitoneum applied to the abdomen. They are contrast-enhanced MR images (enhanced-T1 high-resolution isotropic volume examination (eTHRIVE)). The dataset was created at and owned by The Intervention Center, Oslo University Hospital - Rikshospitalet, Norway.



Medical Imaging, Image Registration, Deformable Image Registration, Medical Image Processing