Cortical Surface Motion Estimation for Brain Shift Prediction. Joldes, G. R., Wittek, A., & Miller, K. January 2010.
Cortical Surface Motion Estimation for Brain Shift Prediction [link]Paper  doi  abstract   bibtex   
In this chapter we present an algorithm for computing the displacement of the exposed surface of the brain during surgery. The motion of the cortical surface is recovered by registering the preoperative surface extracted from high-resolution MRI images to the intraoperative surface acquired during surgery. The recovered motion can then be used for driving a biomechanical model in order to predict the displacement of the internal brain structures, especially the tumor, which can be presented to the surgeon using an image-guided neurosurgery system. Our algorithm combines an image registration method with curvature information associated with the features of the brain surface to perform the non-rigid surface registration. The extracted displacement field can be used directly for driving a biomechanical model, as it does not include any implausible deformations. It also works in cases when the boundaries of the two registered surfaces are not identical, extracting the displacements only for the overlapping regions of the two surfaces. We tested the accuracy of the proposed algorithm using synthetically generated data as well as surfaces extracted from preoperative and intraoperative MRI images of the brain.
@Bookchapter{2010janjoldesmillercortical,
  abstract  = {In this chapter we present an algorithm for computing the displacement of the exposed surface of the brain during surgery. The motion of the cortical surface is recovered by registering the preoperative surface extracted from high-resolution MRI images to the intraoperative surface acquired during surgery. The recovered motion can then be used for driving a biomechanical model in order to predict the displacement of the internal brain structures, especially the tumor, which can be presented to the surgeon using an image-guided neurosurgery system. Our algorithm combines an image registration method with curvature information associated with the features of the brain surface to perform the non-rigid surface registration. The extracted displacement field can be used directly for driving a biomechanical model, as it does not include any implausible deformations. It also works in cases when the boundaries of the two registered surfaces are not identical, extracting the displacements only for the overlapping regions of the two surfaces. We tested the accuracy of the proposed algorithm using synthetically generated data as well as surfaces extracted from preoperative and intraoperative MRI images of the brain.},
  author    = {Joldes, Grand Roman and Wittek, Adam and Miller, Karol},
  booktitle = {Computational Biomechanics for Medicine},
  date      = {2010-01-01},
  doi       = {10.1007/978-1-4419-5874-7_6},
  isbn      = {978-1-4419-5873-0},
  month     = jan,
  publisher = {Springer},
  title     = {Cortical Surface Motion Estimation for Brain Shift Prediction},
  url       = {http://dx.doi.org/10.1007/978-1-4419-5874-7_6},
  year      = {2010},
}
Downloads: 0