Magnetic robotic fetoscope advances life-saving twin surgery with unprecedented precision

In Switzerland, roughly 2,400 twin pregnancies occur each year. When those twins share a placenta—which happens in a subset of these cases—their blood vessels intertwine. The arrangement is not always benign. If blood circulation becomes imbalanced, one fetus can drain blood into the other at a dangerous rate, starving one twin while overloading the other. This condition, called twin-to-twin transfusion syndrome, or TTTS, strikes about twenty to thirty pregnant women annually in Switzerland. It is life-threatening to both children. Until now, the only proven treatment has been surgery—delicate, complex, and performed by only a handful of specialists worldwide.

The operation itself is a feat of precision medicine. Surgeons thread a fetoscope, a thin endoscope designed for work inside the womb, through the mother's abdomen. The instrument carries a camera and a laser fiber. The goal is to identify and seal the blood vessels connecting the two fetuses, restoring balance to their circulation. The problem is mechanical. Current fetoscopes are rigid. Depending on where the placenta sits, some vessels become unreachable. A surgeon's hand, no matter how steady, cannot bend the instrument enough to access them all.

Researchers at ETH Zurich and surgeons at the University of Zurich decided to rebuild the fetoscope from the ground up. They created a flexible version, just 3.2 millimeters in diameter, with magnets embedded in its tip. Three large electromagnets positioned outside the patient's body generate a controllable magnetic field. The magnets in the fetoscope's tip respond to this field the way a compass needle responds to Earth's magnetism, bending and steering the instrument with precision. The team achieved a bend of up to 173 degrees—enough to reach vessels that were previously inaccessible. Michelle Mattille, a postdoctoral researcher and lead author of the work published in Science Robotics, notes that the magnetic field itself poses no danger to the fetus or mother.

But magnetic steering alone was not enough. During minimally invasive surgery, a surgeon sees only a small window of the surgical field through the endoscope's camera. They must memorize the layout of vessels, their paths, their connections—a task that demands years of experience and intense concentration. The new system solves this by stitching together the endoscopic images in real time, creating a panoramic map of the entire surgical area. A surgeon can now point to a target on this two-dimensional map, and the fetoscope navigates there automatically. If the surgeon needs to take manual control, they use a PlayStation game controller. Software translates their input into the corresponding shift in the magnetic field.

The difference in performance is striking. In laboratory tests, participants using the robotic system hit simulated targets far more accurately than with conventional instruments. The robotic platform deviated by roughly 140 micrometers on average—more than four times more precise than comparable existing tools. Mattille observed something unexpected: even the steadiest hand of an experienced surgeon causes the fetoscope to wobble slightly. The robotic platform eliminates that wobble, holding the tip rock-solid at the desired location while the laser coagulates the vessels.

Moving from the laboratory to the real world required a critical test. The team conducted an experiment on a pregnant ewe—the first time they had left the controlled environment behind. They faced the animal's breathing and heartbeat, cloudy amniotic fluid, suspended particles, and a moving fetus. The team spent a year preparing for the three-hour procedure. When they performed it, they successfully replicated the key steps of a TTTS operation under realistic conditions. Mattille describes it as a watershed moment: they were the first group to demonstrate that the procedure could work outside the lab.

The path to human trials is not yet clear. The team must conduct further safety assessments and refine the algorithms to create a system robust enough for clinical use. They are also developing additional assistance functions to support surgeons during procedures. Beyond fetal surgery, the technology could eventually assist in gastroscopy and cystoscopy—procedures where mapping a three-dimensional cavity, rather than a relatively flat surface like the placenta, presents different challenges. For now, the focus remains on TTTS. Although it affects only a small fraction of twin pregnancies, for the families it touches, the stakes could not be higher. A more precise tool, guided by magnetic fields and panoramic vision, could mean the difference between two healthy children and a tragedy.