A newly developed image fusion algorithm between CECT image and CT image: A feasibility study

Cancer cases have been on the rise over the world. Cancer treatment can benefit from an early accurate diagnosis. Percutaneous needle biopsy under the guidance of CT images is the most common method to obtain tumor samples for accurate diagnosis. However, due to the lack of vascular information in the CT images, the biopsy procedure is at great risk, especially for the tumor surrounded by vessels. In this study, a biomechanical model and surface elastic registration-based fusion algorithm were developed to map the vessels from contrast-enhanced CT images of the liver and lung to the corresponded CT image. Radiologists could observe vessels in the CT images during the biopsy procedure so that the risk can be decreased. The developed algorithm was tested through 20 groups of lung data and 16 groups of liver data. The results show that the fusion errors (mean ± standard deviation) were 2.35 ± 0.85, 2.08 ± 0.41, 2.31 ± 0.49, and 2.37 ± 0.62 mm for portal vein, hepatic vein, pulmonary artery, and pulmonary vein, respectively. The accuracy of this method was satisfied in clinical application.