A novel robot for CT-guided bone needle insertion with a rotational drilling and force-feedback speed control mechanism: preliminary evaluation in swine

Abstract Purpose To evaluate the feasibility and accuracy of computed tomography (CT)-guided needle insertion into swine bones by using a novel robotic system capable of rotational drilling. Materials and methods This was an animal experiment using three swine. A remote-controlled robot equipped with a rotational drilling and force-feedback insertion speed control mechanism was developed for bone needle insertion. Using the robot, CT-guided insertion of a 10-gauge bone access needle was attempted in the lumbar vertebrae, ilia, and femora six times each. Needle insertion accuracy was evaluated using the angle error, which is defined as the difference between the predetermined and post-insertion needle angles on axial and sagittal CT images. The time required for needle insertion was measured. The angle error and time required for needle insertion were compared among the bones using the Kruskal–Wallis test. Adverse events were also evaluated. Results Robotic bone needle insertion was successful in all attempts. The median axial and sagittal angle errors were 0.21 and 0.21° for the lumbar vertebrae, 0.32 and 0.13° for the ilia, and 0.65 and 0.25° for the femora, respectively. Axial angle errors were significantly different among the bone types ( p = 0.038). The time required for needle insertion was 23.6, 21.3, and 59.7 s for the lumbar vertebrae, ilia, and femora, respectively. Time was significantly different among bone types ( p = 0.017). No adverse events were observed. Conclusion CT-guided bone needle insertion using a robot equipped with a rotational drilling and force-feedback insertion speed control mechanism was feasible and accurate in swine.