Robotics and augmented reality improve precision in complex jaw surgery

The first, robot‑assisted osteotomy, uses an optical tracking system and a collaborative robotic arm to guide the bone saw. A hand‑eye calibration procedure determines the exact pose of the robot’s end‑effector relative to a tracking marker. Intraoperatively, the surgeon traces the bone surface with an optical probe; the acquired point cloud is registered to the preoperative CT model using the Super4PCS algorithm. The robotic arm then moves a slotted jig to the planned osteotomy plane. The slot physically constrains the saw blade, ensuring precise angles and depths while the tracking marker compensates for any patient micromotion.

“Our system covers the whole chain – from preoperative planning to osteotomy and finally to segment positioning and fixation – within a single, integrated framework,” explains Professor Jian Yang, co‑corresponding author. RAMRS consists of two modules. 

“This gives us angular control without needing an invasive custom guide,” says Professor Jingfan Fan.

The second module, augmented reality‑guided reconstruction (ARR), tackles the challenge of aligning fibular segments to the defective mandible. A quick‑response (QR) marker is attached to the mandible. However, handheld probe picking of the marker’s four corners can be inaccurate due to lever‑arm deformation and hand tremor. The team therefore developed a rotating‑calipers‑based compensation (RCC) method. The algorithm projects the picked points onto a best‑fit plane, finds the minimum‑area enclosing rectangle, and then enforces the known 50 mm side length of the QR marker. “RCC reduces corner displacement by over 25% compared to conventional least‑squares or constrained optimization,” notes Dr. Long Shao. After this refinement, the virtual mandibular reconstruction model is overlaid onto a live video feed, giving the surgeon an intuitive color‑coded projection of where each fibular segment should be placed.

The team evaluated RAMRS on 25 fibula and 25 mandible specimens, then on 8 cadaver legs and 8 cadaver heads to mimic clinical conditions. For fibula osteotomy, RAMRS achieved a mean angular error of 3.19° (SD 1.39°) and a centroid‑distance error of 1.28 mm (SD 0.59 mm). Compared with previously reported methods, this angular error was 9% lower than assisted computer modelling, 42% lower than contour‑registration AR, and 22% lower than combined osteotomy‑reconstruction pre‑shaped plates. For mandible osteotomy, the feature‑point distance error was 1.18 mm (SD 0.59 mm) and the volumetric error just 4.34% – substantially better than the image‑guided sagittal saw (8.55%).

In the reconstruction phase, the RCC method reduced AR fusion error to 1.26 mm (SD 0.42 mm) – a 42‑48% improvement over the two baseline methods. The final reconstructed mandible showed a condylar distance error of 1.38 mm, a mandibular‑angle distance error of 1.36 mm, and an angular error of 3.62°. All metrics were competitive or superior to published clinical series using pre‑bent plates or CAD‑CAM guides.

The cadaver experiments confirmed that the system remains accurate even in the presence of soft tissue. Osteotomy errors stayed below 2 mm, and reconstruction errors remained small despite the reduced bony surface available for registration.

This is a major step forward because most previous work stopped at plastic phantoms. We have validated RAMRS under conditions that closely resemble real surgery – including soft‑tissue retraction and limited exposure.”

Professor Tao Zhang, clinical collaborator

The authors acknowledge that intraoperative blood contamination can still affect marker tracking and AR overlay. Future work will integrate more robust image‑enhancement algorithms and add distance‑warning mechanisms to protect vital structures such as the facial artery. Regulatory preparation and standardized workflow studies are also planned.

“Our vision is a seamless, ‘see‑and‑cut, see‑and‑place’ workflow where the robot handles the geometrically demanding osteotomy and AR provides real‑time visual guidance for reconstruction,” concludes Professor Yang. “RAMRS demonstrates that combining robotics and augmented reality can improve both precision and usability in complex maxillofacial surgery.”

Authors of the paper include Sifan Cao, Jingfan Fan, Long Shao, Qing Sun, Tao Xu, Danni Ai, Tianyu Fu, Deqiang Xiao, Hong Song, Tao Zhang, and Jian Yang.

This article was originally published on news-medical