Measurement error in statistical models of shape

Authors:
Damian J. J. Farnell;Andrew Pickles;Christopher Roberts
Affiliations:
Health Methodology Research Group, School of Community-Based Medicine, Jean McFarlane Building, University Place, University of Manchester, Manchester M13 9PL, United Kingdom and Division of Mathe ...;Health Methodology Research Group, School of Community-Based Medicine, Jean McFarlane Building, University Place, University of Manchester, Manchester M13 9PL, United Kingdom and Institute of Psyc ...;Health Methodology Research Group, School of Community-Based Medicine, Jean McFarlane Building, University Place, University of Manchester, Manchester M13 9PL, United Kingdom
Venue:
Computer Methods and Programs in Biomedicine
Year:
2011

Citing 7
Cited 0

Active shape models—their training and application

Computer Vision and Image Understanding
Automatic Interpretation and Coding of Face Images Using Flexible Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Active Appearance Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Toward Automatic Simulation of Aging Effects on Face Images

IEEE Transactions on Pattern Analysis and Machine Intelligence
Vertebral shape: automatic measurement with dynamically sequenced active appearance models

MICCAI'05 Proceedings of the 8th international conference on Medical image computing and computer-assisted intervention - Volume Part II
Detecting Reduced Bone Mineral Density From Dental Radiographs Using Statistical Shape Models

IEEE Transactions on Information Technology in Biomedicine
Reliability Estimation for Statistical Shape Models

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.01

Visualization

Abstract

Active shape models (ASMs) are popular and sophisticated methods of extracting features in (especially medical) images. Here we analyse the error in placing ASM points on the boundary of the feature. By using replications, a corrected covariance matrix is presented that should reduce the effects of placement error. We show analytically and via simulations that the cumulative variability for a given number of eigenvalues retained in principal components analysis (PCA) ought to be reduced by increasing levels of point-placement error. Results for predicted errors are in excellent agreement with the set-up parameters of two simulated shapes and with anecdotal evidence from the trained experts for real data taken from the OSTEODENT project. We derive an equation for the reliability of placing the points and we find values of 0.79 and 0.85 (where 0=bad and 1=good) for the two clinical experts for the OSTEODENT data. These analyses help us to understand the sources and effects of measurement error in shape models.