Changzhou Jinse Medical Technology Co., Ltd.
2D/3D image registration is particularly important in medical imaging, and it becomes more popular in academy and industry. The objective of this project is to explore key technologies of 2D/3D image registration, addressing issues of image domain differences, 2D image distortion and low processing efficiency. Three solutions are thus proposed. First, a novel domain adaptation method is proposed to solve the issue of image domain difference. Second, a new image calibration method based on a 3D deformation model is proposed to efficiently reduce image distortion. Third, an efficient 2D/3D image registration method based on two-stage registration is proposed to solve the slow registration processing issue. Based on these proposed algorithms, we will conduct feasibility studies on X-ray and CT image registration. The high-performance of the registration algorithms can offer substantial benefits across various domains. In research, it facilitates advanced biomedical studies and algorithm development. In healthcare, it enhances diagnostic accuracy, improves surgical planning, and optimizes treatments. Economically, it leads to cost savings and improved quality of life. Overall, the advancements of 2D/3D image registration have the potential to impact both individual lives and society.
