Dans le contexte des sciences archéologiques, des images tridimensionnelles issues de scanners tomodensitométriques sont segmentées en régions d’intérêt afin d’en faire une analyse. Ces objets virtuels sont souvent utilisés dans le but d’effectuer des mesures précises. Une partie de ces analyses nécessite d’extraire la surface des régions d’intérêt. Cette thèse se place dans ce cadre et vise à améliorer la précision de l’extraction de surface. Nous présentons dans ce document nos contributions : tout d’abord, l’algorithme du HMH pondéré dont l’objectif est de positionner précisément un point à l’interface entre deux matériaux. Appliquée à une extraction de surface, cette méthode pose des problèmes de topologie sur la surface résultante. Nous avons donc proposé deux autres méthodes : la méthode du HMH discret qui permet de raffiner la segmentation d’objet 3D, et la méthode du HMH surfacique qui permet une extraction de surface contrainte garantissant l’obtention d’une surface topologiquement correcte. Il est possible d’enchainer ces deux méthodes sur une image 3D pré-segmentée afin d’obtenir une extraction de surface précise des objets d’intérêt. Ces méthodes ont été évaluées sur des acquisitions simulées d’objets synthétiques et des acquisitions réelles d’artéfacts archéologiques. / In the context of archaeological sciences, 3D images produced by Computer Tomography scanners are segmented into regions of interest corresponding to virtual objects in order to make some scientific analysis. These virtual objects are often used for the purpose of performing accurate measurements. Some of these analysis require extracting the surface of the regions of interest. This PhD falls within this framework and aims to improve the accuracy of surface extraction. We present in this document our contributions : first of all, the weighted HMH algorithm whose objective is to position precisely a point at the interface between two materials. But, applied to surface extraction, this method often leads to topology problems on the resulting surface. So we proposed two other methods : The discrete HMH method which allows to refine the 3D object segmentation, and the surface HMH method which allows a constrained surface extraction ensuring a topologically correct surface. It is possible to link these two methods on a pre-segmented 3D image in order to obtain a precise surface extraction of the objects of interest These methods were evaluated on simulated CT-scan acquisitions of synthetic objects and real acquisitions of archaeological artefacts.

Optimal surface finding (OSF), a graph-based optimization approach to image segmentation, represents a powerful framework for medical image segmentation and analysis. In many applications, a pre-segmentation is required to enable OSF graph construction. Also, the cost function design is critical for the success of OSF. In this thesis, two issues in the context of OSF segmentation are addressed. First, a robust model-based segmentation method suitable for OSF initialization is introduced. Second, an OSF-based segmentation refinement approach is presented. For segmenting complex anatomical structures (e.g., lungs), a rough initial segmentation is required to apply an OSF-based approach. For this purpose, a novel robust active shape model (RASM) is presented. The RASM matching in combination with OSF is investigated in the context of segmenting lungs with large lung cancer masses in 3D CT scans. The robustness and effectiveness of this approach is demonstrated on 30 lung scans containing 20 normal lungs and 40 diseased lungs where conventional segmentation methods frequently fail to deliver usable results. The developed RASM approach is generally applicable and suitable for large organs/structures. While providing high levels of performance in most cases, OSF-based approaches may fail in a local region in the presence of pathology or other local challenges. A new (generic) interactive refinement approach for correcting local segmentation errors based on the OSF segmentation framework is proposed. Following the automated segmentation, the user can inspect the result and correct local or regional segmentation inaccuracies by (iteratively) providing clues regarding the location of the correct surface. This expert information is utilized to modify the previously calculated cost function, locally re-optimizing the underlying modified graph without a need to start the new optimization from scratch. For refinement, a hybrid desktop/virtual reality user interface based on stereoscopic visualization technology and advanced interaction techniques is utilized for efficient interaction with the segmentations (surfaces). The proposed generic interactive refinement method is adapted to three applications. First, two refinement tools for 3D lung segmentation are proposed, and the performance is assessed on 30 test cases from 18 CT lung scans. Second, in a feasibility study, the approach is expanded to 4D OSF-based lung segmentation refinement and an assessment of performance is provided. Finally, a dual-surface OSF-based intravascular ultrasound (IVUS) image segmentation framework is introduced, application specific segmentation refinement methods are developed, and an evaluation on 41 test cases is presented. As demonstrated by experiments, OSF-based segmentation refinement is a promising approach to address challenges in medical image segmentation.