Pierre-Frederic Villard
Laboratoire lorrain de recherche en informatique et ses applications
Campus scientifique - BP 239 - 54506 Vandoeuvre-les-Nancy Cedex
France
Computer-based simulations of respiration can be found in different areas. The level of fidelity that needs to be reached depends on the application domain. In radiotherapy the model accuracy is the first consideration while in computer animation for movies a high degree of fidelity is not essential. For medical training simulators there is a tradeoff between accuracy and speed: i) the perception should be realistic enough to train medical students on cases that are close to the reality and ii) the computation has to be performed in realtime to integrate haptics and graphics.
We present how to simulate respiration within three different application contexts: treatment planning in radiotherapy,training simulators for ultrasound-guided liver biopsy and for on-line help during fluoroscopy-guided interventional radiology
Anatomy and physiology of the respiration system will be reviewed, focusing on each organ of interest, which depends on the procedure. For each context, the respiration modeling technique will be introduced in order to fit with the application requirement.
Depending on the medical application various organs need to be modeled. In the case of simulators it should be possible to vary the type of breathing: tidal breathing, hyperventilation, diaphragmatic or thoracic breathing, etc. Hence, we chose to use a simulated online respiration model, instead of a pre-computed model. We divided the thoracic organs and the viscera into five categories: i) static rigid organs, ii) rigid organs with rotational motion, iii) rigid organs with translational motion, iv) deformable organs, and v) deformable organs with internal contraction.
A method based on an optimisation technique called \u201cevolutionary strategy\u201d has been designed to estimate the parameters the respiration model. This model is adaptable to account for patient\u2019s specificities. The aim of the optimisation algorithm is to finely tune the model so that it accurately fits real patient datasets. Our algorithm is fully automatic and adaptive. A compound fitness function has been designed to take into account for various quantities that have to be minimized (here topological errors of the liver and the diaphragm geometries).
Final results of these different studies will be presented using videos, clinical validation studies, and user feedbacks.