Share Email Print

Proceedings Paper

Training sample reduction through model feature selection in anatomical model development
Author(s): Tod S. Levitt; Marcus W. Hedgcock M.D.; Vera Michele Shadle; Scott E. Johnston; D. N. Vosky
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Model-based reasoning techniques complement traditional image processing and pattern recognition to compensate for the ambiguities inherent in medical imagery data. In model- based reasoning, higher-order image features, such as inferred surface boundaries, are matched against parametric models of anatomy. The instantiated model can be used to predict approximate locations of other anatomical image features for further segmentation processing. The inferential power of this approach is limited by the accuracy with which anatomical models capture population statistics. One possible problem with a model-based approach is that it requires a large training samples to develop models. We show that selection of quasi- invariant features greatly reduces the population sample required to accurately model population distribution of features. Quasi-invariant features include certain ratios, angles and other functions of directly measurable image features that constrain each others values. When one observable is measured in the image, the likely values and locations of others are relatively constrained. For example, the ratios of lengths of phalanges has small variance across the population of all humans compared to the variance of each bone length individually. An experiment is presented on a population of 90 radiographs that supports this approach to segmentation for hand anatomy measures.

Paper Details

Date Published: 1 July 1992
PDF: 6 pages
Proc. SPIE 1654, Medical Imaging VI: PACS Design and Evaluation, (1 July 1992); doi: 10.1117/12.60314
Show Author Affiliations
Tod S. Levitt, San Francisco VA Medical Ctr. and Univ. of California/San Fr (United States)
Marcus W. Hedgcock M.D., San Francisco VA Medical Ctr. and Univ. of California/San Fr (United States)
Vera Michele Shadle, San Francisco VA Medical Ctr. (United States)
Scott E. Johnston, United Programmers (United States)
D. N. Vosky, San Francisco VA Medical Ctr. (United States)

Published in SPIE Proceedings Vol. 1654:
Medical Imaging VI: PACS Design and Evaluation
R. Gilbert Jost M.D., Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?