Nonlinear filtering and fast ray tracing of 3D image data. Improving visualization in medical imaging

This article deals with the problem of nonlinear filtering of magnetic resonance (MR) tomograms based on geometry-driven diffusion and speeding up the visualization of volumetric data by ray tracing. The authors present the basic motivation of this research, which was recently carried out in the Institute of Measurement Science. Locally adaptive diffusion conductivity, effective iteration stopping criterion, and fast ray tracing algorithms are discussed. Also, the results of computer experiments are described and the improvement of 3-D visualization is demonstrated.