[Virtual training in head- and neurosurgery]

Abstract

Background: Surgical skill requires training at close-to-reality scenarios. Conventional procedures (practical exercises, anatomical or animal cadavers) are increasingly in conflict with objective parameters (costs of OR, quality management, cadavers availability and cost). Surgical procedures can be covered by using Virtual Surgical Reality (VSR).

Methods: We examined the principle of a VSR system and evaluated the results with 30 probands by the example of modified radical mastoidectomy and ventriculocisternostomy. Probands were divided in experienced (A) and non-experienced surgeons (B). The protocol included time of surgery, collisions and fatal injuries at altogether 15 passages (10 passages - break of 14 days - 5 passages). Additionally the Medical Level of Trust (LOT) describes the confidence into the surgical scenario and thus the quality by volume illustration, texture and haptic data feedback to the user. We used a numeric scale between 0 and 100 and the starting point of 50.

Results: The learning effect can be confirmed for both virtual scenarios. The mastoid scenario reaches a total confidence index LOT of 75. The ventricle scenario is evaluated with a total confidence index of 84. The necessary time for the opening of the antrum is reduced from 15.0 minutes (group A) and 22.5 minutes (group B) around 37 % to 9, 5 (group A) and around 56 % to 10.0 minutes (group B). The virtual haptic result was evaluated positive in both scenarios.

Conclusions: VSR systems have the potential to revolutionize surgical training. All surgical experienced probands evaluated the VSR-scenario as near-to-reality. "Suspension of Disbelief" is the major condition for effective virtual reality training systems.