In a laboratory in Montreal, a millimeter-sized robot made of biocompatible rubber bends and glides through transparent channels of flowing liquid, guided by invisible magnetic fields—a breakthrough that could one day spare patients from the risks of conventional surgery. Researchers at Concordia University have developed an artificial intelligence-assisted platform featuring small, magnetically guided soft robots capable of navigating the body's most delicate blood vessels to locate and remove life-threatening clots deep within neurovascular pathways.
The innovation matters because treating blood clots in complex vessels currently relies on traditional catheter-based procedures that carry real risks: surgeons can accidentally damage or perforate fragile vessel walls during intervention. These tiny magnetic soft robots, tethered to the tips of conventional catheters and surgical wires, offer a safer alternative. By attaching the robots to existing surgical tools, doctors can steer them with far greater precision and control, then withdraw them with minimal trauma to surrounding tissue.
The robotic platform is built around a permanent magnet mounted on a six-axis robotic arm that controls the robot's movements with remarkable accuracy. What sets this system apart from earlier magnetic robotic designs is its sophistication: rather than relying on bulky electromagnets and simple open-loop controls, the Concordia team developed a closed-loop system that continuously measures and monitors the robot's position in real time. The researchers created two deep-learning models—one that predicts how the soft robot will behave under changing magnetic forces, gravity, and fluid-flow conditions inside the body, and another that detects the robot's shape and precise tip position from high-speed camera images. This constant feedback allows the system to adapt and maintain accuracy even as conditions shift.
The proof is in the testing. Researchers conducted a series of in vitro experiments using transparent fluidic channels designed to mimic the vascular environment, complete with flowing liquid that simulates blood flow. The results were striking: compared to standard approaches, the new system reduced tracking effort by as much as 77 percent while requiring significantly less control effort overall. Even more impressively, the closed-loop control system consistently outperformed conventional methods, demonstrating greater accuracy, stability, and resilience against disturbances caused by flowing fluid.
"These tiny magnetic soft robots are attached to the tips of conventional catheters and surgical wires, allowing surgeons to steer the tethered robot toward obstructions, perform the intervention and bring it back with lower risk," says Ramin Sedaghati, a professor in the Department of Mechanical, Industrial and Aerospace Engineering at the Gina Cody School of Engineering and Computer Science and one of the paper's authors. He notes that wireless magnetic fields "open up a lot of applications for health care and minimally invasive surgery."
The work is genuinely multidisciplinary, weaving together materials design, robotics, computational and experimental mechanics, artificial intelligence, and control systems. Alireza Moezi, lead author and Ph.D. 2026 graduate now a postdoctoral fellow at McGill University, worked alongside Subhash Rakheja and Sedaghati. Their findings appear in the journal Smart Materials and Structures, signaling that this technology has moved beyond theory into validated prototype territory. As magnetic-guided robotics mature, surgeons could soon offer patients a path to treating dangerous clots with unprecedented safety and precision.