Evolutionary algorithms for medical simulations: a case study in minimally-invasive vascular interventions

To obtain the expertise to correctly perform minimally– invasive vascular interventions thorough training is required. Training using simulation systems are increasingly becoming an accepted methodology. Recently, a minimally–invasive vascular intervention simulation (MIVIS) system has been developed. At the heart of this system lies an optimization problem to be solved repeatedly. In this paper, we investigate the advantages and disadvantages of using an evolutionary algorithm (EA) to solve the optimization problem instead of a problem–specific first–order analytical approximation algorithm. The results show that the use of the EA as optimization algorithm is favorable. A substantial reduction in time can be obtained while the RMS error associated with the simulation result differs only slightly. Categories and Subject Descriptors

• Type: Conference
• Year: 2005
• Where: GECCO
• Authors: Peter A. N. Bosman, Tanja Alderliesten
• Keywords: GECCO 2005, Optimization Problem, Vascular Intervention, Vascular Intervention Simulation