Surgical practice is shifting towards robotic-assisted minimally invasive surgery (RMIS), with procedures being performed through small incisions or natural orifices in order to minimise trauma to the patient, with the technological support of robots.

RMIS uses telemanipulation setups to control the surgical instruments and also to introduce computer-assisted interventions (CAI) in the operation, such as overlaying intra- and pre-operative imaging onto the video feed, for enhanced visualisation of structural and functional anatomical information underneath the visible tissue surface.

The practical implementation of CAI systems is critically dependent on performing a hand-eye calibration that links the robot and the surgical camera coordinate systems. Currently, available hand-eye calibration methods are not accurate enough in the context of RMIS. One of the main conditions for an accurate hand-eye calibration is to acquire images of a calibration target with a wide range of camera motions that fully explore all 6DoF of the problem, while due to mechanical constraints surgical robots limit the camera motion to only 4DoF (3 in rotation and 1 in translation).

In our recent work, we have developed a new approach to determine the hand-eye transformation in such a constrained setting, by first estimating the remote centre of motion in the camera reference frame and then using this information to simplify the classic hand-eye formulation. This suggests that a more convenient path towards accurate hand-eye calibration in robotic surgery is through correctly modelling its motion constraints rather than allowing the robot to freely move in a calibration phase (this is neither practical nor possible in many cases).

Relevant publications

  • Pachtrachai, K., Vasconcelos, F., Dwyer, G., Hailes, S., & Stoyanov, D. (2019). Hand-eye calibration with a remote centre of motion. IEEE Robotics and Automation Letters, 1. doi:10.1109/lra.2019.2924845
  • Pachtrachai, K., Vasconcelos, F., Dwyer, G., Pawar, V., Hailes, S., & Stoyanov, D. (2018). CHESS - Calibrating the Hand-Eye Matrix with Screw constraints and Synchronisation. IEEE Robotics and Automation Letters. doi:10.1109/LRA.2018.2800088