Does virtual reality training make sense in interventional radiology? Training skill-, rule- and knowledge-based behavior.

Abstract

To be able to meet the fast-growing demands of interventional radiology and endovascular therapy, many more interventional radiologists have to be trained in procedures that are becoming more complex. Until recently, fellows obtained their first basic skills training during diagnostic angiography procedures. This traditional method of training is disappearing because other diagnostic methods became available which raises the question “How can our future interventional radiologists be trained?” Training on animals, bench models (skill boxes) and virtual reality (VR) training has been suggested as a solution for this problem [1–5]. In many countries, animal experiments have been confronted with a growing resistance because animals are too expensive to be used as a full-scale training tool. With the use of bench models, objective assessment of performance is difficult to obtain. Training on simulators does not have these problems and seems therefore to be a good alternative. However, the effectiveness and efficiency of VR trainers in interventional radiology (IR) have not yet been addressed. In aviation, human behavior models have been implemented to develop and to evaluate new training methods [6, 7]. These models are useful for specifying the objectives, needs and means of training methods [8–10]. This paper introduces Rasmussen’s model of human behavior and applies it to IR. This model will be used to discuss the capabilities and limitations of current VR for IR training in relation to the training needs.