Automated Task Load Detection with Electroencephalography: Towards Passive Brain-Computer Interfacing in Robotic Surgery.

Thorsten O. Zander,Kunal Shetty,Romy Lorenz,Daniel Richard Leff,Laurens R. Krol,Ara Darzi,Klaus Gramann,Guang-Zhong Yang

DOI: https://doi.org/10.1142/s2424905x17500039

2017-01-01

Abstract:Automatic detection of the current task load of a surgeon in the theatre in real time could provide helpful information, to be used in supportive systems. For example, such information may enable the system to automatically support the surgeon when critical or stressful periods are detected, or to communicate to others when a surgeon is engaged in a complex maneuver and should not be disturbed. Passive brain–computer interfaces (BCI) infer changes in cognitive and affective state by monitoring and interpreting ongoing brain activity recorded via an electroencephalogram. The resulting information can then be used to automatically adapt a technological system to the human user. So far, passive BCI have mostly been investigated in laboratory settings, even though they are intended to be applied in real-world settings. In this study, a passive BCI was used to assess changes in task load of skilled surgeons performing both simple and complex surgical training tasks. Results indicate that the introduced methodology can reliably and continuously detect changes in task load in this realistic environment.