Title
Registration Graphs: A Language for Modeling and Analyzing Registration in Image-Guided Surgery

Abstract
Computer-assisted surgical systems must bring diagnostic images, surgical plans, patient anatomy, surgical tools, robots, vision systems, and other components into accurate alignment with one another. Multi-step registration procedures have been devised, which are difficult to analyze, or even to describe concisely.

A method for diagramming registration strategies and procedures makes descriptions straightforward and simplifies analysis. A notation that uses a graph theoretic framework consisting of two primary elements: features, representing objects (or parts of objects); and links, representing measurement actions is introduced. Connectivity properties of the resulting registration graph are readily determined by inspection.

To provide quantitative as well as qualitative analysis of a registration strategy, a model for propagating measurement uncertainties through the graph is presented. The model involves the first and second statistical moments, mean and covariance. Methods for combining the means and covariances of serial as well as parallel measurements are shown.

The rules and algorithms governing the registration graph and uncertainty analysis are implemented as a C++ library and Windows program. A case study of a registration strategy verifies that the library and program are effective, and shows that registration is more than just a "final alignment between tool and patient."

Source: Ph.D. dissertation, Northwestern University, December 1998

Title
Registration and Immobilization in Robot-Assisted Surgery

Abstract
Robotic systems for computer assisted surgery involve both tools and techniques which are new to the surgical arena. Registration and immobilization in particular are key problems. Registration is the spatial alignment of the coordinate frames of the robot, an anatomic object (e.g., a bone), and the preoperative plan (a computer model). Immobilization is necessary to maintain that alignment. We discuss various approaches to registration and immobilization, and solutions appropriate for an orthopedic surgical system.

Source: Journal of Image Guided Surgery 1 (2) 1995

Title
Diagramming Registration Connectivity and Structure

Abstract
Computer-assisted surgical systems must bring preoperative diagnostic images, surgical plans, patient anatomy, surgical tools, robots, and other components into accurate alignment with one another. Multi-step registration procedures have been devised, which are difficult to analyze, or even to describe concisely.

Here we introduce a notation for diagramming registration strategies and procedures, and apply it to a spectrum of current methods. The notation uses a graph-theoretic framework consisting of two elements: nodes, representing objects; and links, representing actions. Connectivity properties of the resulting registration graph are readily determined by inspection.

Using the registration graph one may confirm the validity of registration, check if frameless surgery is possible, list serial sources of error, determine which objects must be rigidly fixated, determine the existence and extent of redundant registration pathways, identify structurally identical registration methods, and devise and describe new registration architectures.

Source: IEEE Engineering in Medicine and Biology, May 1995