Automatic Deformable Mr-ultrasound Registration for Image-guided Neurosurgery

Registration of preoperative MR images and Ultrasound images are used for the purpose of accuracy in detecting the tumor. Image Guided Neurosurgery Systems (IGNS) can be used to track surgical tools with respect to the preoperative Magnetic Resonance (MR) images, brain tissue movement during surgery invalidates the image-topatient mapping and thus reduces the effectiveness of using preoperative images for intraoperative surgical guidance. The movement of brain is caused by biochemical factors, and is referred to as brain shift. The various technique used for this purpose are RaPTOR, which is used to reduce the mean target registration errors from patients. The artificial neural network is used to set the data base about the patient’s registered medical images. The Bivariate correlation ratio is used to correlate the images. The Gaussian blur normalized mutual information is used to allow automatic US-MRI registration in few seconds. Image segmentation is used to segment the tumor image. The US and MRI images registered are compared tat which techniques reduces the mean target registration error. RaPTOR calculate local correlation ratio (CR) values on small patches and adds the CR values to form a global cost function. It is therefore constant to large amounts of spatial intensity in homogeneity. We also propose a novel outlier suppression technique based on the direction of the RaPTOR gradients. RaPTOR is validated on tracked US images and MR of neurosurgery. Deformable registration of the US and MR images acquired respectively pre-operation and post-resection is of significant clinical significance, but challenging due to, among others, the large amount of missing correspondences among the two images. This work is also novel in that it performs automatic registration of this challenging dataset. for validating the results, we manually locate corresponding anatomical landmarks in the MR and US images of tumor resection in brain surgery. Keywords— Brain surgery, IGNS, image guided neurosurgery, intra operative ultrasound, nonrigid registration, online database, validation database.