Skip to Main content Skip to Navigation

Segmentation et recalage d'images échographiques par utilisation de connaissances physiologiques et morphologiques

Gelu Ionescu 1
1 TIMC-GMCAO - Gestes Medico-chirurgicaux Assistés par Ordinateur
TIMC - Techniques de l'Ingénierie Médicale et de la Complexité - Informatique, Mathématiques et Applications, Grenoble - UMR 5525
Abstract : Echographical imaging could potentially play a major role in the field of Computer Assisted Surgery (CAS). For doctors and surgeons to make full use of this tool in planning and executing surgical operations, they also need user-friendly automatic software based on fast, precise and reliable algorithms. The main goal of this thesis is to take advantage of the segmentation/registration duality to extract the relevant information from echographical images. This information will allow the precise and automatic registration both of anatomical structures contained in the pre-operative model and of per-operative data contained in echographical images. The result of registration will be further used to guide a computer-assisted tool. In the first part we propose différent methods for filtering, segmentation and calibration of echographical images. The development of fast, precise and reliable algorithms is emphasized. The second part concerns the segmentation-registration duality and the corrections of elastic deformations of soft tissues. High-level segmentation algorithms for echographical images were developed. They are based on results of low-level segmentation, a priori anatomical knowledge as well as on information provided by the pre-operative model. The third part deals with detailed descriptions of applications and interpretation of results. The cases studied include : screwing inside the vertebral pedicles, iliosacral screwing, prostatic radiotherapy and puncture of pericardial effusion. Future developments for this approach are discussed.
Document type :
Complete list of metadatas
Contributor : Thèses Imag <>
Submitted on : Tuesday, March 2, 2004 - 4:35:23 PM
Last modification on : Tuesday, November 24, 2020 - 4:18:04 PM
Long-term archiving on: : Friday, April 2, 2010 - 9:14:10 PM


  • HAL Id : tel-00005189, version 1



Gelu Ionescu. Segmentation et recalage d'images échographiques par utilisation de connaissances physiologiques et morphologiques. Ingénierie biomédicale. Université Joseph-Fourier - Grenoble I, 1998. Français. ⟨tel-00005189⟩



Record views


Files downloads