Clinical Significance of Creative 3D-Image Fusion Across [CT + MR] Modalities Based on Approach of Characteristic Co-Registration
Matthew Jian‐Qiao Peng, Yong Hu, Xiangyang Ju, Daozhang Cai · Journal of Medical Imaging and Health Informatics · 2016
Abstract: Since three-dimensional (3D) hybrid detector 【 CT+MR 】 is not integrated well currently, this study aims to investigate a registration scheme for two-dimensional (2D) hybrid based on characteristic localization to achieve 3D-fusion from the images of CT and MR as a whole. A cubic oriented proposal of ―9-point & 3-plane ‖ for coregistration designs were verified to be geometrically practical. Human internal-feature points were sorted to combine with preselected external-feature points for matching process through 3D-reconstruction and virtual-dissection. By following the procedures of feature-extraction and image-mapping, the processes of ―picking points to form plane ‖ and ―picking planes for segment ‖ were executed. Ultimately, image-fusions were implemented at the real-time workstation Mimics based on auto-fuse techniques so called ―information exchange ‖ and ―Signal Overlaying‖. A complementary 3D-image across 【 CT+MR 】 modalities, which simultaneously presents anatomic structures of hard-tissue and soft-tissue, was created with a detectable-rate of 70%, this is equivalent to detectable-rate of 【 PET+CT 】 or 【 PET+MR 】 with no statistically significant difference. Our approach of ―9-point & 3-plane ‖ offers a fresh idea for integration of digital imaging in mathematic consideration, and it facilitates a 3D vision that isn’t functional yet for 2D hybrid imaging. This exploration is practical to those small hospital that are unable to afford expensive hybrid equipment.