A MINIATURE magnetic robot utilizing a novel oloid shape has enabled high-resolution 3D ultrasound imaging deep within the gastrointestinal tract, offering transformative potential for early cancer detection through virtual biopsies, according to a collaborative study led by the University of Leeds.
Colorectal cancer, a leading cause of cancer-related deaths globally, currently requires invasive tissue sampling and weeks-long diagnostic delays. This research introduces a 21-mm-diameter oloid-shaped magnetic flexible endoscope (OME) equipped with a 28 MHz ultrasound probe, capable of non-invasively generating 3D cross-sectional tissue reconstructions resembling physical biopsies. The oloid’s unique geometry—formed by intersecting perpendicular circles—enables a six-degree-of-freedom rolling motion under external magnetic control, overcoming previous limitations of cylindrical robots. Tested in artificial colons and pig models, the OME successfully performed autonomous navigation and identified lesions with 150-μm resolution, equivalent to histological analysis.
The study, published in Science Robotics, demonstrated the OME’s ability to capture 3D ultrasound scans deep in the gut, facilitating real-time diagnosis during procedures. In preclinical trials, the system achieved 98% navigation accuracy and reduced procedural time by 40% compared to traditional colonoscopy. Clinical adoption could eliminate the 1–3-week wait for biopsy results, enabling same-session treatment. The team aims to commence human trials by 2026, building on ongoing trials of a related robotic colonoscopy platform. Future applications may extend to esophageal and gastric cancers, with potential for ultrasound-triggered drug delivery. Clinically, this technology could address gender disparities in colonoscopy outcomes, as the OME’s autonomy reduces operator-dependent challenges.
Reference
Greenidge N et al. Harnessing the oloid shape in magnetically driven robots to enable high resolution ultrasound imaging. Science Robotics. 2025;DOI:10.1126/scirobotics.adq4198.
