A HIFU Localization System Based on Two-Dimensional Multimodality Image Registration

In the interactive image-guided high intensity focused ultrasound (HIFU) tumor therapy, fast and precise target localization is very important for treatment planning. Current localization techniques under ultrasound image guidance can not satisfy the requirement of precision. This paper presents a method of target localization based on 2D multimodality image registration. CT images prior to therapy were introduced into the system to aid tumor localization. According to the characteristics of the image series scanning mode in abdominal HIFU surgery, both CT and ultrasound images were re-sampled as slices parallel to the treatment plane, and with the CT/ultrasound image registration of the same anatomical layer, a spatial transform of the clear tumor contour from CT to the ultrasound coordinate system can be accomplished. A HIFU surgery localization system has been designed and preliminarily simulated.

[1]  Guy Marchal,et al.  Multimodality image registration by maximization of mutual information , 1997, IEEE Transactions on Medical Imaging.

[2]  Yang-Ming Zhu,et al.  Likelihood maximization approach to image registration , 2002, IEEE Trans. Image Process..

[3]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[4]  T. Loupas,et al.  Noise reduction in ultrasonic images by digital filtering. , 1987, The British journal of radiology.

[5]  Yazhu Chen,et al.  Non-rigid Registration Algorithm Based on Modally Controlled Free Form Deformation , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[6]  Yazhu Chen,et al.  A diffusion stick method for speckle suppression in ultrasonic images , 2004, Pattern Recognit. Lett..