### Abstract

A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate remarkably high accuracy in superposition of images from different subjects, thus enabling very precise localization of morphological characteristics in population studies. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e. a set of geometric moment invariants that is defined on each voxel in an image, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure. This is a fundamental deviation of our method, referred to as HAMMER, from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e. suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting features that have distinct attribute vectors, thus drastically reducing ambiguity in finding correspondence. A number of experiments in this paper have demonstrated excellent performance.

Original language | English |
---|---|

Title of host publication | Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis |

Editors | L. Staib |

Pages | 29-36 |

Number of pages | 8 |

Publication status | Published - 2001 Dec 1 |

Externally published | Yes |

Event | Workshop on Mathematical Methods in Biomedical Image Analysis MMBIA 2001 - Kauai, HI, United States Duration: 2001 Dec 9 → 2001 Dec 10 |

### Other

Other | Workshop on Mathematical Methods in Biomedical Image Analysis MMBIA 2001 |
---|---|

Country | United States |

City | Kauai, HI |

Period | 01/12/9 → 01/12/10 |

### Fingerprint

### ASJC Scopus subject areas

- Analysis

### Cite this

*Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis*(pp. 29-36)

**HAMMER : Hierarchical attribute matching mechanism for elastic registration.** / Shen, Dinggang; Davatzikos, Christos.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis.*pp. 29-36, Workshop on Mathematical Methods in Biomedical Image Analysis MMBIA 2001, Kauai, HI, United States, 01/12/9.

}

TY - GEN

T1 - HAMMER

T2 - Hierarchical attribute matching mechanism for elastic registration

AU - Shen, Dinggang

AU - Davatzikos, Christos

PY - 2001/12/1

Y1 - 2001/12/1

N2 - A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate remarkably high accuracy in superposition of images from different subjects, thus enabling very precise localization of morphological characteristics in population studies. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e. a set of geometric moment invariants that is defined on each voxel in an image, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure. This is a fundamental deviation of our method, referred to as HAMMER, from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e. suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting features that have distinct attribute vectors, thus drastically reducing ambiguity in finding correspondence. A number of experiments in this paper have demonstrated excellent performance.

AB - A new approach is presented for elastic registration of medical images, and is applied to magnetic resonance images of the brain. Experimental results demonstrate remarkably high accuracy in superposition of images from different subjects, thus enabling very precise localization of morphological characteristics in population studies. There are two major novelties in the proposed algorithm. First, it uses an attribute vector, i.e. a set of geometric moment invariants that is defined on each voxel in an image, to reflect the underlying anatomy at different scales. The attribute vector, if rich enough, can distinguish between different parts of an image, which helps establish anatomical correspondences in the deformation procedure. This is a fundamental deviation of our method, referred to as HAMMER, from other volumetric deformation methods, which are typically based on maximizing image similarity. Second, in order to avoid being trapped by local minima, i.e. suboptimal poor matches, HAMMER uses a successive approximation of the energy function being optimized by lower dimensional energy functions, which are constructed to have significantly fewer local minima. This is achieved by hierarchically selecting features that have distinct attribute vectors, thus drastically reducing ambiguity in finding correspondence. A number of experiments in this paper have demonstrated excellent performance.

UR - http://www.scopus.com/inward/record.url?scp=0035700327&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035700327&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0035700327

SP - 29

EP - 36

BT - Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis

A2 - Staib, L.

ER -