Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection

Sajjad Baloch, Christos Davatzikos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Groupwise registration and statistical analysis of medical images are of fundamental importance in computational anatomy, where healthy and pathologic anatomies are compared relative to their differences with a common template. Accuracy of such approaches is primarily determined by the ability of finding perfectly conforming shape transformations, which is rarely achieved in practice due to algorithmic limitations arising from biological variability. Amount of the residual information not reflected by the transformation is, in fact, dictated by template selection and is lost permanently from subsequent analysis. In general, an attempt to aggressively minimize residual results in biologically incorrect correspondences, necessitating a certain level of regularity in the transformation at the cost of accuracy. In this paper, we introduce a framework for groupwise registration and statistical analysis of biomedical images that optimally fuses the information contained in a diffeomorphism and the residual to achieve completeness of representation. Since the degree of information retained in the residual depends on transformation parameters such as the level of regularization, and template selection, our approach consists of forming an equivalence class for each individual, thereby representing them via nonlinear manifolds embedded in high dimensional space. By employing a minimum variance criterion and constraining the optimization to respective anatomical manifolds, we proceed to determine their optimal morphological representation. A practical ancillary benefit of this approach is that it yields optimal choice of transformation parameters, and eliminates respective confounding variation in the data. Resultantly, the optimal signatures depend solely on anatomical variations across subjects, and may ultimately lead to more accurate diagnosis through pattern classification.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume6918
DOIs
StatePublished - 2008
Externally publishedYes
EventMedical Imaging 2008 - Visualization, Image-Guided Procedures, and Modeling - San Diego, CA, United States
Duration: Feb 17 2008Feb 19 2008

Other

OtherMedical Imaging 2008 - Visualization, Image-Guided Procedures, and Modeling
CountryUnited States
CitySan Diego, CA
Period2/17/082/19/08

Fingerprint

Equivalence classes
Image analysis
Statistical methods
Electric fuses
Pattern recognition

Keywords

  • Computational anatomy
  • Deformation based morphometry
  • Groupwise registration
  • Statistical image analysis
  • Voxel-based morphometry

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Baloch, S., & Davatzikos, C. (2008). Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 6918). [691805] https://doi.org/10.1117/12.769947

Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection. / Baloch, Sajjad; Davatzikos, Christos.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6918 2008. 691805.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Baloch, S & Davatzikos, C 2008, Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 6918, 691805, Medical Imaging 2008 - Visualization, Image-Guided Procedures, and Modeling, San Diego, CA, United States, 2/17/08. https://doi.org/10.1117/12.769947
Baloch S, Davatzikos C. Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6918. 2008. 691805 https://doi.org/10.1117/12.769947
Baloch, Sajjad ; Davatzikos, Christos. / Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6918 2008.
@inproceedings{8d6ccad6f47a479abf769674954397db,
title = "Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection",
abstract = "Groupwise registration and statistical analysis of medical images are of fundamental importance in computational anatomy, where healthy and pathologic anatomies are compared relative to their differences with a common template. Accuracy of such approaches is primarily determined by the ability of finding perfectly conforming shape transformations, which is rarely achieved in practice due to algorithmic limitations arising from biological variability. Amount of the residual information not reflected by the transformation is, in fact, dictated by template selection and is lost permanently from subsequent analysis. In general, an attempt to aggressively minimize residual results in biologically incorrect correspondences, necessitating a certain level of regularity in the transformation at the cost of accuracy. In this paper, we introduce a framework for groupwise registration and statistical analysis of biomedical images that optimally fuses the information contained in a diffeomorphism and the residual to achieve completeness of representation. Since the degree of information retained in the residual depends on transformation parameters such as the level of regularization, and template selection, our approach consists of forming an equivalence class for each individual, thereby representing them via nonlinear manifolds embedded in high dimensional space. By employing a minimum variance criterion and constraining the optimization to respective anatomical manifolds, we proceed to determine their optimal morphological representation. A practical ancillary benefit of this approach is that it yields optimal choice of transformation parameters, and eliminates respective confounding variation in the data. Resultantly, the optimal signatures depend solely on anatomical variations across subjects, and may ultimately lead to more accurate diagnosis through pattern classification.",
keywords = "Computational anatomy, Deformation based morphometry, Groupwise registration, Statistical image analysis, Voxel-based morphometry",
author = "Sajjad Baloch and Christos Davatzikos",
year = "2008",
doi = "10.1117/12.769947",
language = "English (US)",
isbn = "9780819471024",
volume = "6918",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Anatomical equivalence class based complete morphological descriptor for robust image analysis and abnormality detection

AU - Baloch, Sajjad

AU - Davatzikos, Christos

PY - 2008

Y1 - 2008

N2 - Groupwise registration and statistical analysis of medical images are of fundamental importance in computational anatomy, where healthy and pathologic anatomies are compared relative to their differences with a common template. Accuracy of such approaches is primarily determined by the ability of finding perfectly conforming shape transformations, which is rarely achieved in practice due to algorithmic limitations arising from biological variability. Amount of the residual information not reflected by the transformation is, in fact, dictated by template selection and is lost permanently from subsequent analysis. In general, an attempt to aggressively minimize residual results in biologically incorrect correspondences, necessitating a certain level of regularity in the transformation at the cost of accuracy. In this paper, we introduce a framework for groupwise registration and statistical analysis of biomedical images that optimally fuses the information contained in a diffeomorphism and the residual to achieve completeness of representation. Since the degree of information retained in the residual depends on transformation parameters such as the level of regularization, and template selection, our approach consists of forming an equivalence class for each individual, thereby representing them via nonlinear manifolds embedded in high dimensional space. By employing a minimum variance criterion and constraining the optimization to respective anatomical manifolds, we proceed to determine their optimal morphological representation. A practical ancillary benefit of this approach is that it yields optimal choice of transformation parameters, and eliminates respective confounding variation in the data. Resultantly, the optimal signatures depend solely on anatomical variations across subjects, and may ultimately lead to more accurate diagnosis through pattern classification.

AB - Groupwise registration and statistical analysis of medical images are of fundamental importance in computational anatomy, where healthy and pathologic anatomies are compared relative to their differences with a common template. Accuracy of such approaches is primarily determined by the ability of finding perfectly conforming shape transformations, which is rarely achieved in practice due to algorithmic limitations arising from biological variability. Amount of the residual information not reflected by the transformation is, in fact, dictated by template selection and is lost permanently from subsequent analysis. In general, an attempt to aggressively minimize residual results in biologically incorrect correspondences, necessitating a certain level of regularity in the transformation at the cost of accuracy. In this paper, we introduce a framework for groupwise registration and statistical analysis of biomedical images that optimally fuses the information contained in a diffeomorphism and the residual to achieve completeness of representation. Since the degree of information retained in the residual depends on transformation parameters such as the level of regularization, and template selection, our approach consists of forming an equivalence class for each individual, thereby representing them via nonlinear manifolds embedded in high dimensional space. By employing a minimum variance criterion and constraining the optimization to respective anatomical manifolds, we proceed to determine their optimal morphological representation. A practical ancillary benefit of this approach is that it yields optimal choice of transformation parameters, and eliminates respective confounding variation in the data. Resultantly, the optimal signatures depend solely on anatomical variations across subjects, and may ultimately lead to more accurate diagnosis through pattern classification.

KW - Computational anatomy

KW - Deformation based morphometry

KW - Groupwise registration

KW - Statistical image analysis

KW - Voxel-based morphometry

UR - http://www.scopus.com/inward/record.url?scp=44949127249&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44949127249&partnerID=8YFLogxK

U2 - 10.1117/12.769947

DO - 10.1117/12.769947

M3 - Conference contribution

SN - 9780819471024

VL - 6918

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -