ROBNEWS
A magnetic module at the centre of a silicone body is remagnetised to switch between locomotion, cutting tissue, releasing drugs, gripping samples, and localised heating. The five tasks are demonstrated in lab tests only, not in living tissue.
A 4.4-millimetre soft robot built at NTU Singapore can switch between five surgical functions in under a second by remagnetising a single magnetic module at its centre, the team reports. The result was published in Advanced Materials, with a preprint on arXiv at 2509.15610. The work, led by Associate Professor Lum Guo Zhan of the university's School of Mechanical and Aerospace Engineering, frames the device as a step toward a single millimetre-scale tool that could, in principle, navigate inside a body and perform several tasks without being swapped out.
The device is a magnetic soft microrobot: a millimetre-scale body made of silicone-based elastomers embedded with magnetic particles, steered by external magnetic coils rather than onboard batteries or motors. The NTU team used two common medical silicones, PDMS and Ecoflex, loaded with 5-micrometre magnetic microparticles. The body is moulded once. What changes between tasks is the magnetisation pattern inside it.
The mechanism behind the five functions is one reprogrammable core. A weak external field can magnetise, demagnetise, and remagnetise that core along different axes, each axis corresponding to a different mode of movement or interaction with tissue. The team's reported switch time is under a second. The five modes the NTU press release and the Robohub summary describe are: locomotion across a surface, cutting biological tissue, on-demand drug release, gripping and storing a tissue sample, and localised heating.
The advance is in stacking those modes into one body without retooling. Most prior magnetic microrobots do one or two things well. A capsule that can be steered to a tumour site and heated is useful; an elastomer that can be steered and can cut a biopsy sample is a different device. The NTU paper's claim is that one moulded silicone body, plus a programmable magnetic field, can act as both, and three more besides, in sequence. That is what "5-in-1" is shorthand for in this paper.
That claim is so far a bench-top one. The team's experiments used tissue-mimicking gels and excised biological samples; the paper, the NTU release, and the Robohub write-up report no in-vivo or human data. The magnetic coils that drive the device are bench-top laboratory units, not the kind of image-guided electromagnetic steering system a clinical interventional suite would need to guide a tool through a body. The team's stated long-term goal is in-body navigation to a targeted site for minimally invasive treatment, but that is a roadmap, not a demonstrated capability.
The gap between cutting tissue in a dish and cutting tissue inside a patient is not just scale. It is selective actuation, fail-safe behaviour, sterilisation, biocompatibility over hours rather than minutes, and the regulatory path that any energy-delivering surgical device has to clear. A device that heats on demand has to be selectively heatable; a device that grips tissue has to let go on cue; a device that releases drugs has to do so under controlled conditions, not on its own schedule. None of those are visible in a five-function list.
The honest read of the result is narrower than the press cycle suggests. The paper demonstrates that a single reprogrammable magnetic core can drive several distinct surgical modes in a millimetre-scale soft body, in vitro, on a bench. That is a real engineering result, and it is the result to anchor the story on. The five-function list is what the core enables, not a forecast of when the device reaches a clinic. A peer-reviewed in-vivo safety study, or a replication in a tissue phantom under image guidance, would be the first concrete step toward the team's stated application.