Skip to Main content Skip to Navigation
Theses

Exploring new numerical methods for the simulation of soft tissue deformations in surgery assistance

Abstract : This thesis addresses the problem of soft tissue simulation for augmented reality applications in liver surgery assistance and, more specifically, the implementation of a non-rigid registration pipeline to be used by the medical staff to generate interactive deformations of a patient specific liver three-dimensional virtual representation. A formal physics-based framework is first defined and used as the basis for the construction of a biomechanical model capable of producing realistic deformations. Four basic requirements guided the development of the model: accuracy, speed, stability and simplicity of implementation. Meshless and immersed-boundary methods are both considered as alternatives to the traditional finite element method. A formal non-rigid registration algorithm is finally documented and tested with real-life scenarios. A comparison with new and rising machine learning and neural network solutions is also provided.
Complete list of metadata

https://hal.inria.fr/tel-03130643
Contributor : Jean-Nicolas Brunet Connect in order to contact the contributor
Submitted on : Wednesday, February 3, 2021 - 5:19:37 PM
Last modification on : Wednesday, March 17, 2021 - 8:18:30 AM

File

thesis_jean-nicolas_brunet_fin...
Files produced by the author(s)

Identifiers

  • HAL Id : tel-03130643, version 1

Collections

Citation

Jean-Nicolas Brunet. Exploring new numerical methods for the simulation of soft tissue deformations in surgery assistance. Computer Science [cs]. Université de Strasbourg, 2020. English. ⟨tel-03130643⟩

Share

Metrics

Les métriques sont temporairement indisponibles