Laparoscopic Surgery Simulator

Gilles Debunne, Pierre-Olivier Agliati, Marie-Paule Cani , Fabrice Neyret,
Mathieu Desbrun, Jean-Christophe Lombardo



Aims and Scope

  Motivation: Non-invasive surgery
  Aim: Simulator for surgeons training.
  Difficulties: Real-time deformable model for the liver, force feedback, realistic rendering.


Overview

  We rely on the graphics hardware for performing real-time collision detection between the liver model and the surgical tools.
  Real-time animation is achieved thanks to an adaptive model that automatically concentrates computations where needed.
  Realistic visualisation is provided thanks to a multi-pass rendering. The liver surface is covered with undistorted cells thanks to a new texturing techniques that non-periodically maps a few precomputed patterns on triangles that tile the surface. Animated textures are used for enhancing the changes of the liver's surface aspect under the tools actions.
Details on realistic rendering




Results

Static collision detection (14 K)
Dynamic collision detection (17 K)
Liver deformation during a contact with the tool (128 K)
Realistic rendering of the surface aspect of the liver


Publications and related works

  Adaptive Simulation of Soft Bodies in Real-Time .
G. Debunne, M. Desbrun, M.P. Cani & A. Barr.
Computer Animation 2000. Philadelphia, USA, May 2000.

  Interactive multiresolution animation of deformable models.
G. Debunne, M. Desbrun, A. Barr & M.P. Cani.
10th Eurographics Workshop on Computer Animation and Simulation (CAS'99), Milano, sep 1999.

  Real-time Collision Detection for Virtual Surgery.
J.C. Lombardo, M.P. Cani & F. Neyret.
Computer Animation'99, Geneva, may 1999.

  Pattern-based Texturing Revisited.
F. Neyret & M.P. Cani.
SIGGRAPH'99, aug 1999.

Action incitative INRIA AISIM