The role of computed tomography data in the design of a robotic magnetically-guided endoscopic platform

Abstract

Seventy cases of computed tomography (CT) data were selected from The Cancer Imaging Archive to aid in the design of a magnetically-guided capsule endoscope platform. According to the distance between the large bowel and the abdomen skin measured on CT images, the supine position showed the advantage in reducing the distance between internal and external magnets; thus, we used the supine position as the detection posture for magnetically-guided capsule colonoscopy. From the selected 70 cases of CT data, we chose 30 supine cases with different waistline lengths to extract the centerline of the large bowel as the motion trajectory of the capsule. We analyzed these trajectories to define the workspace of the capsule and simulated the motion of the capsule inside the large bowel. The distance between the large bowel and the abdomen surface, which can be regarded as the minimum distance of the internal and external magnets, was calculated and used as the reference for designing the magnet. Based on the abdomen volume and the workspace of the capsule, we provided a minimum workspace of the external magnet and designed an auxiliary degree of freedom for the manipulator to control the capsule to pass through the large bowel.