MRI is the prime imaging modality for the neurosurgery cases we support. The images are acquired prior to surgery without a need for special landmarking strategies. To process the imagery, a wide range of methods (e.g., [25,21,16,19,22]) have been applied to the segmentation problem, i.e. identifying different tissue types in medical imagery. Our current approach to segmentation uses an automated method to initially segment into major tissue classes while removing gain artifacts from the imager [25,13], then uses operator driven interactive tools to refine this segmentation. This latter step primarily relies on 3D visualization and data manipulation techniques to correct and refine the initial automated segmentation. The segmented tissue types include skin, used for registration, and internal structures such as brain, tumor, vessels, and ventricles. These segmented structures are processed by the Marching Cube algorithm [14] to construct isosurfaces and support surface rendering for visualization.
The structural models of patients constructed using such methods can be augmented with functional information. For example, functional MRI methods or transcranial magnetic stimulation methods [7] can be used to identify motor or sensory cortex. This data can then be fused with the structural models [26] to augment such models. In each case, segmentation produces information in a local coordinate frame, which must be merged together. We currently use a registration method based on Mutual Information [26] to do this. The result is an augmented, patient-specific, geometric model of relevant structural and functional information.