Klinische Neurophysiologie 2004; 35 - 216
DOI: 10.1055/s-2004-832128

Visualizing the Dynamics of Structural Brain Change

C Preul 1, M Tittgemeyer 2, DY von Cramon 3
  • 1Leipzig
  • 2Leipzig
  • 3Leipzig

Introduction: A morphometric assessment of the dynamics of structural brain change relies on a precise and informative visualization. Time-series examinations with magnetic resonance imaging (MRI) allow the assessment and monitoring of changes in intra-individual brain structures that are altered by pathological processes or surgical intervention. Three examples resemble structural alterations in the context of selected diseases: 1) Ventricular adjustment after surgical relief of elevated intracranial pressure in internal hydrocephalus corresponds to highly dynamic changes. 2) Intraoperative brain shift of the subthalamic nucleus as the target structure for deep brain stimulation in the treatment of Parkinson's disease demands a high precision in monitoring. 3) Long-term changes such as structural degeneration in Alzheimer's disease necessitate a subtle follow-up. Methods: The evaluation of the structural change follows a multi- step approach: Upon reorientation to the bi-commissural plane, images are processed for grey-scale normalization and intensity inhomogeneities. Rigid registration is applied to align the images followed by a fuzzy means algorithm to segment tissue types. A region-growing algorithm was adapted to the assessment of ventricular volume. Cortical surfaces are generated as smooth triangular meshes. Non-rigid registration is employed to monitor residual differences between the images, which reflect morphological changes. The resulting vector field is 1:1 superimposed to the MR image to allow a direct measurement of structural change. Results and Conclusion: The capability of the deformation-based analysis to visualize tissue changes of high (hydrocephalus) and low temporal (Alzheimer's disease) dynamics will be illustrated in detail as well as subtle changes due to brain shift in the deep nuclei. The deformation-based analysis is a valuable tool in the interpretation of MR images. Such an analysis is applicable to any disease or condition where a careful observation of tissue changes is of particular importance for a reliable monitoring of the course of a disease or the success of a neurosurgical intervention.