Abstract
MRI could not achieve real-time navigation for High Intensity Focused Ultrasound(HIFU) due to its slow refresh rate. Therefore, we proposed an ultrasound(US) guided HIFU navigation system. To avoid potential imaging error, a rotation scan is used to construct 3D ultrasound image for navigation. However discontinuity emerges in the cross-sectional images of the reconstructed volume data. The reason is that images taken from different directions with different speckle patterns associated, intersect around the center of the volume. The discontinuity may cause difficulty in segmentation process and blur diagnostic information. An algorithm which applies a mask to the acquired images before 3D reconstruction is presented to resolve this problem. This method gives all the images taken from all different directions appropriate weight based on their spatial relation. With this method, cross-sectional image with more uniform speckle pattern is obtained. This work improves the continuity of US image used for navigation and could ease the task of automatic target detection. In the future, we will continue to develop US-guided navigation system to achieve more efficient HIFU treatment.
