Textured laser range scanning and registration of the cortical surface

A novel method for intra-operative data acquisition using a laser range scanner is presented for applications in model-updated image-guided surgery. The scanner produces a point cloud containing true-color encoded range data of its field-of-view. This allows characterization of intra-operative conditions using both intensity and geometry information. Furthermore, a rigid-body registration algorithm is developed to incorporate the scanner's unique data into image-guided surgery applications. The algorithm builds on iterative closest point transforms and mutual information. Phantom registration experiments have shown submillimetric accuracies using the scanner and the algorithm. In this paper we present a set of qualitative results from a neurosurgical case.