Colorectal cancer is one of the most frequently diagnosed cancers and a leading cause of cancer-related death, and survival depends strongly on early detection. Conventional colonoscopy is the diagnostic standard, but it is invasive, uncomfortable, and carries risks such as bowel perforation. Wireless capsule endoscopy is a less invasive alternative, but current capsules cannot be steered and rely on peristalsis, risking incomplete coverage of the colon.
This project works towards steering the capsule magnetically using an external rotating permanent magnet, which requires knowing the position of both the capsule and the magnet. Hall sensors are commonly used for this kind of tracking, typically placed in an evenly distributed grid. In this work, the sensor positions are optimised for tracking both magnets, avoiding the positions and orientations where an even grid would otherwise yield low measurement sensitivity.
The optimised layouts are used for simultaneous tracking of both magnets. The magnetic field of the rotating magnet that drives the capsule is far stronger than the capsule's own and dominates the sensor measurements. This project estimates the positions of both magnets directly and simultaneously from the same sensor measurements, rather than tracking the capsule only after removing the actuator's field from them.