Image processing in stereotactic planning: volume visualization and image registration.

Abstract

Radiation treatment planning (RTP), historically an image-intensive discipline and one of the first areas in which three-dimensional (3D) information from imaging was clinically applied, has become ever more critically dependent on accurate 3D identification of target and non-target organs with the advent of conformal radiation therapy, stereotactic radiotherapy, and radiosurgery. In addition to the interactive display of wire-frame or shaded surface models of anatomical objects, proposed radiation beams, and calculated dose distributions, use may also be made of direct visualization of relevant anatomy from image data. Although anatomical analysis of dose distributions is an essential component of radiation treatment design which requires geometric definition (i.e., segmentation) of all volumes of interest, geometric targeting with optimization based on 3D anatomical information is frequently performed as a separate step independent of dose calculations. For this early step in the planning process, and certainly for diagnostic purposes, visualization without explicit segmentation may be a useful additional capability. Additionally, we suggest that direct visualization of high contrast targets such as arteriovenous malformations (AVM) from computerized tomography (CT) or magnetic resonance (MR) angiography may serve as a tool for target volume delineation. Frequently, information from multiple modalities or multiple scans of the same modality is important in planning a given case. In such instances, image registration may prove useful to allow synthesis of information from multiple sources into a single consistent coordinate frame. Numerous image registration methods are available, each with characteristic strengths and weaknesses.