Improving the reliability of shape comparison by perturbation

Authors:
Y. Jiang;E. Edmiston;F. Wang;H. P. Blumberg;X. Papademetris;L. H. Staib
Affiliations:
Department of Diagnostic Radiology, Yale University, New Haven, CT;Psychiatry, Yale University, New Haven, CT;Psychiatry, Yale University, New Haven, CT;Department of Diagnostic Radiology and Psychiatry, Yale University, New Haven, CT;Department of Diagnostic Radiology and Biomedical Engineering, Yale University, New Haven, CT;Department of Diagnostic Radiology and Biomedical Engineering and Electrical Engineering, Yale University, New Haven, CT
Venue:
ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
Year:
2009

Citing 5
Cited 0

Bagging predictors

Machine Learning
Performance Issues in Shape Classification

MICCAI '02 Proceedings of the 5th International Conference on Medical Image Computing and Computer-Assisted Intervention-Part I
Unsupervised Learning of an Atlas from Unlabeled Point-Sets

IEEE Transactions on Pattern Analysis and Machine Intelligence
Shape Registration in Implicit Spaces Using Information Theory and Free Form Deformations

IEEE Transactions on Pattern Analysis and Machine Intelligence
A study of hippocampal shape difference between genders by efficient hypothesis test and discriminative deformation

MICCAI'07 Proceedings of the 10th international conference on Medical image computing and computer-assisted intervention - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

Shape comparison is a key scenario in morphometric study, where registration is often involved and found to be unreliable: different registrations can lead to different shape differences. This paper proposes a generic scheme applicable to most registration methods, to reduce this unreliability. It perturbs the registration processes by feeding them with resampled shape groups, and then aggregates the results to yield the final result. This scheme can be simplified for pairwise registration methods to reduce the computation. Experiments are conducted on both synthetic and biomedical shapes using different registration methods, which demonstrate its effectiveness.