Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo

Siming Bayer M. Sc.

Researcher in the Image Fusion (IMF) group at the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg

Brain Shift Compensation
  • Brain shift is the change of the position and shape of the brain during a neurosurgery procedure e.g. an open skull surgery. This intraoperative brain deformation limits the use of neuroanatomical overlays that were produced prior to the surgery. In the practice, brain shift can be observed as a relative motion of the brain tissue with respect to the skull after craniotomy. The combination of various factors, e.g. cerebrospinal fluid leakage, gravity, edema, tumor mass e ect, brain parenchyma resection or retraction and administration ofosmotic diuretics leads to the intraoperative brain shift phenomenon. This phenomenon was shown to be an important source of error that needs to be considered during a neurosurgery procedure.

    Basically, the brain shift can be estimated and compensated with intraoperatively generated Magnetic Resonance (MR) images. However, MR scanners are very expensive and their imaging process is very time consuming due to patient transfers. Other medical imaging modalities e.g. DynaCT and Digital Subtraction Angiography (DSA) are more suitable for the interventional setting. They are interventionally established and the anatomical structures of the whole head can be reconstructed within several seconds. But detailed anatomical information about the soft tissues, e.g. brain or tumor tissue cannot be captured with these modalities. For these reasons, a reliable and robust real time method, which estimates and compensates the intraoperative brain shift to ensure the accuracy during neurosurgery procedure, will be developed in this research project.

    With non-rigid registration algorithms, the brain shift can be measured on fast acquisitions of intraoperative 3D DSA (or DynaCT) images. This transformation is then used to match the preoperative MR images on the intraoperative data. In literature, non-rigid registration of pre- and intraoperative MR images is well studied. However, as mentioned before, the acquisition of intraoperative MRI is time consuming and very expensive, and therefore is not suitable for an interventional setting such as an open skull operation. Since the anatomical and morphological information about the soft tissue on the MR images is necessary for the surgeon and a real time solution to compensate the brain shift is desirable, we propose in this research project a new approach to estimate and correct the intraoperative brain deformation indirectly via 3D DSA (or DynaCT) images. This approach overcomes the image acquisition problem with intraoperative MR and provides the surgeon with the necessary information during a neurosurgery procedure in real time.