Multimodal Neuronavigation Surgery Of Motor Area Brain Tumor With Functional Magnetic Resonance Imaging

Part 1 The experimentalstudies of multimodality images registration accuracy for multimodal neuronavigation Objective Relying on a single imaging modality during image guidance surgery can sometimes omit visualization information that another modality could provide. Integration of multimodal medical images offers surgeon the ability to combine multiple image exams of various imaging modalities from a single patient for individualized and optimal navigation visualization. Registration accuracy is the key marker for the evaluation of multimodality images integration. In this experimental study, an anthropomorphic head phantom was used to access the registration accuracy of commercial StealthMerge Image Registration Software which would be applied in the following multimodal neuronavigation surgical planning. And the correlation facts with the registration accuracy were analyzed. Clinical Material and Methods # A plastic-lucite Peg-Board Phantom (34 , Medtronic Surgical Navigation Technologies, Louisville, USA) filled with water and containing tomography (CT) and magnetic resonance (MR) visible targets was modified to evaluate the accuracy of multimodality image registration and fusion software. For MR imaging, the peg-board cavity was filled with water. For CT scanning, it remained empty except for air. Lucite rods were distributed as a matrix on thebottom of the square base of the phantom. On top of each rod, there was a conical.shaded bore with its apex at the center of the free tip, which be used as anatomic landmark models for image registration. Different sets of phantom image studies were obtained: CT, T1-weighted MR imaging, T2-weighted MR imaging and single shot spin-echo echoplanar imaging (SE EPI). The studies were divided into five sets, referring to the imaging modalities. Set A was combination CT and MR imaging by using AutoMerge registration method; Set B was combination CT and MR imaging by using PointMerge registration method; Set C was combination T1-weighted and T2-weighted MR imaging by using AutoMerge registration method; Set D was combination 2mm and 5mm thickness --8 第一部分多影像融合神经导航技术的实验研究 T1-weighted MR imaging by using AutoMerge registration method; Set E was combination SE EPI and T1-weighted MR imaging by using AutoMerge registration method respectively. The registration accuracy of each anatomiclandmark models was measured one by one. The general linear model (GLM) was used for statistical analyze. The multiple factors ANOVA was used to examined the significant difference between each group. LSD method was applied to uncover the main effects between each couple of sets (a=0.05). The multiple variances correlation...