Accepted:
----
| |
/vol02/vajdic/ | © Copyright 1995 | |||
| Volume 01 | Received: Accepted: |
---- ---- |
|||
|
AI and Medical Imagery: Strategy and Evaluation of Inexact Relational Matching
Stevan M. Vajdic, Michael J. Brooks, Andrew Downing and Henry E. Katz |
|
| Abstract | |
| Matching/fusion strategies are discussed and a set of functions for evaluation of relational inexact matching is introduced. They are implemented in the domain of 2D/3D medical imagery and are based on Artificial Intelligence (AI) paradigms? in particular, rule-based knowledge representation and tree search. The 2D reference and target images are selected from 3D sets and are segmented into non-touching and non-overlapping regions, using iterative thresholding and/or knowledge about the anatomical regional shapes of human organs. Selected image region attributes and their inter-relationships (relations) are calculated. Region matches are obtained using a tree search, and the error is minimised by evaluating a "goodness of matching" function based on similarities of region attributes/relationships. Once the matched regions are found and the spline geometric transform is applied to regional centres of gravity, images are ready for fusion into a single 2D and/or 3D image of higher clarity. | |
| Full Text |
|
|
Multimedia Links (none) Reference Links (none) Citation Reference |
Get viewers for PS & PDF   |
