Camara-Rey, Oscar (2003) Non-linear registration of thoracic and abdominal CT and 18-FDG whole-body emission PET images: methodological study and application in clinical routine. PhD thesis Signal et Images, ENST - TSI Traitement du Signal et des Images, ENST.
Full text available as:
|
|
Abstract
The aim of this work is to implement an algorithm to achieve a robust, fast enough and good quality registration of thoracic and abdominal CT and 18-FDG whole-body emission PET images. The proposed registration methodology is based on the incorporation of prior anatomical information in an intensity-based non-linear registration algorithm. This incorporation is performed in an explicit way, by initializing the intensity-based registration stage with the solution obtained by a registration of corresponding anatomical surfaces segmented through a hierarchically ordered set of anatomy-specific rules. Deformations are modeled in both registration steps by means of a FFD model. Mutual Information is used as the similarity criterion at the grey-level registration step. Registration errors provided by the visual assessment protocol we have designed are less than 1cm on lungs, heart, liver and kidneys structures but up to 1.5cm on the stomach.
| Item Type: | PhD Thesis (PhD) |
|---|---|
| Thesis Supervisor: | Bloch, Isabelle |
| Date: | December 2003 |
| Board of examiners: | Vandermeulen, Dirk and Grangeat, Pierre and Villanueva, Juan José and Heitz, Fabrice and Cinquin, Philippe |
| Ecole Doctorale: | ED 130 INFORMATIQUE, TELECOMMUNICATIONS ET ELECTRONIQUE (EDITE) |
| Discipline: | Signal et Images |
| Collection (Fonds): | ENST |
| Institution: | ENST |
| Department: | ENST - TSI Traitement du Signal et des Images |
| Subjects: | 2. Information and Communication Sciences and Technologies 1. Mathematics and Applications |
| Uncontrolled Keywords: | Nonlinear registration, Ct, Pet, Thoracic and abdominal images, Free Form Deformations, Mutual Information, Gradient Vector Flow, Deformable models, Hierarchical segmentation, Spatial relationships, Anatomical constraints, Recalage non lineaire, Tdm, Tep, Images thoraciques et abdominales, Deformations a Forme Libre, Information Mutuelle, Gradient Vector Flow, Modeles deformables, Segmentation hierarchique, Relations spatiales, Contraintes anatomiques, Puesta en correspondencia no lineal, Imagenes toracicas y abdominales, Deformacions de Forma Libre, Informacion Mutua, Gradient Vector Flow, Modelos deformables, Segmentacion jerarquica, Relaciones espaciales, Restricciones anatomicas, Posada en correspondencia no lineal, Tac, Tep, Imatges toraciques i abdominals, Deformacions de Forma Lliure, Informacio Mutua, Gradient Vector Flow, Models deformables, Segmentacio jerarquica, Relacions espacials, Restriccions anatomiques |
Table of content
-Part 1. Application context
*Chapter 1. Medical Framework
*Chapter 2. Registration theory
-Part 2. Proposed methodology
*Chapter 3. Initial tests and proposed methodology
*Chapter 4. Structure segmentation
*Chapter 5. Structure registration
*Chapter 6. Grey-level registration
*Chapter 7. Validation of the registration methodology
*Chapter 8. Results
*Chapter 9. Conclusions and perspectives
-Part 3. Appendices
*Appendix A. Find-Judas registration application
*Appendix B. Medical information about cancer
*Appendix C. Details of segmentation methods
*Appendix D. Semi-interactive segmentation of tumors
*Appendix E. Description of brain internal structures by means of spatial relations for MR image segmentation
| ID Code: | 903 |
|---|---|
| Deposited By: | Oscar Camara Rey |
| Deposited On: | 03 February 2005 |
Repository Staff Only: edit this item