Robust robot-camera calibration for robotized Transcranial Magnetic Stimulation

Type of publication:  Artikel
Zeitschrift: International Journal of Medical Robotics and Computer Assisted Surgery
Band: 7
Nummer: 4
Jahr: 2011
Monat: Dezember
Seiten: 414-422
Notiz: Transcranial Magnetic Stimulation
DOI: 10.1002/rcs.411
Abriss: Transcranial Magentic Stimulation (TMS) is a method to painlessly and noninvasively stimulate parts of the brain. For robotized TMS, the magnetic coil is placed on the patient's head by a robot. As the robotized TMS system requires tracking of head movements, robot and tracking camera need to be calibrated. Conventionally, such calibration is done by tracking the robot's end effector, i.e., moving the robot into different poses and measuring the position and orientation of a marker attached to the end effector. Hence, robot motions can be related to the respective motion in camera coordinates. Generally, solving this problem is called hand-eye-calibration, and several methods for precise estimation of the related transformations exist. However, for robotized TMS in a clinical setting such calibration is required frequently, especially since robot and tracking system are partially mobile. Mounting / unmounting a marker to the end effector and moving the robot into different poses is rather impractical. Moreover, if either system is moved during treatment, recalibration is required. To overcome this limitation, we propose to directly track link three of the articulated arm. Using forward kinematics and a constant marker transform, the calibration can be performed instantly. This robust online calibration greatly enhances the system's user-friendliness as the paramount aim for all medical robotized systems is easy and safe usability in clinical routine and research. Furthermore, we can recompute the transformation throughout the treatment, i.e., we can constantly check for changes and potentially interrupt the treatment. Our experimental results indicate an accuracy similar to standard hand-eye calibration approaches. For the robotized TMS system, the overall accuracy was 2.2mm using the proposed approach, which is more than twice the accuracy of other navigated TMS systems.
Nutzerfelder: file={ress_11.pdf:ress_11.pdf:PDF}
Autoren: Richter, Lars
Ernst, Floris
Schlaefer, Alexander
Schweikard, Achim
  • ress_11.pdf