Researchers at The Chinese University of Hong Kong have developed an artificial intelligence-driven magnetic tracking system designed to locate and guide millimeter-scale wireless medical robots while they are being moved by magnetic fields.
The work addresses a persistent challenge in miniature medical robotics: tracking a tiny wireless device inside the body at the same time that strong magnetic fields are being used to actuate it.
The findings were published in Science Robotics. The system remains pre-clinical, but the researchers say it could help move magnetic microrobots closer to practical minimally invasive diagnosis and treatment.
Tracking during magnetic actuation
Miniature magnetic robots can potentially reach narrow or difficult anatomical spaces without wires or conventional surgical instruments. Their size and wireless operation, however, make it difficult to determine exactly where they are once they move inside tissue.
The CUHK team, led by Professor Zhang Li from the Department of Mechanical and Automation Engineering, developed an AI-based approach to suppress interference caused by the magnetic actuation system.
The platform achieved millimeter-level localization accuracy at penetration depths of up to 10 centimeters in biological tissue. Tracking information was then fed back into the actuation system to enable closed-loop navigation rather than relying only on open-loop magnetic commands.
Large-animal validation
The researchers tested the technology in a goat model, navigating miniature magnetic robots through the spinal subarachnoid space and the bile duct.
The experiments took place in an unshielded clinical environment with multiple sources of magnetic interference. The estimated trajectories closely matched X-ray observations, according to CUHK.
That validation is significant because laboratory tracking performance can deteriorate in realistic environments where medical equipment and surrounding infrastructure introduce additional interference.
Clinical translation remains the next step
The research does not yet establish clinical use in humans. Further work would be required around device design, safety, workflow integration, regulatory approval and the reliability of navigation across different anatomical settings.
Still, real-time localization is one of the technical requirements for moving miniature robots from controlled experiments toward procedures where a clinician needs to know where a device is and how it is moving.
The development adds to a wider push to combine AI, robotics and biomedical engineering in Asia. TNGlobal recently reported on Mayo Clinic Laboratories’ investment in Pathology Asia and LifeStrands Genomics as advanced healthcare technologies move closer to regional clinical workflows.
For magnetic microrobots, the next milestone will be showing that the tracking and navigation system can remain accurate, safe and practical across a broader range of clinically relevant procedures before human use is considered.
Mayo Clinic Laboratories invests in Pathology Asia and LifeStrands Genomics