Diagnostic Requirements for Efficient, Adaptive Robotic Surgery Training
Abstract
Robot-assisted surgery training is shifting towards simulation-based training. Challenges that accompany this shift are high costs, working hour regulations and the high stakes aspects of the surgery domain. Adaptive training could be a possible solution to reduce the problems. First, an adaptive system needs diagnostic data with which the system can make an action selection. A scoping literature search was performed to give an overview of the state of the research regarding diagnostic requirements. Diagnostic metrics should be (a) useful for formative and not only summative assessment of trainee progress, (b) valid and reliable, (c) as nonintrusive as possible for the trainee, (d) predictive of future performance in the operating theater (e) explanatory, and (f) suitable for real-time assessment of trainee's learning state. For a more in-depth understanding, further research is needed into which simulator parameters can be used as diagnostic metrics that can be assessed in real-time. A possible framework for adaptive training systems is discussed, and future research topics are presented.
