In context
In 2023, minimally invasive spine surgery was gaining traction for its benefits of smaller incisions and faster recovery, yet surgeons still relied heavily on intraoperative X-ray and 2D images, which limited precision and increased radiation exposure. Augmented reality (AR) was emerging as a promising tool to overlay 3D medical imagery onto the patient, but practical clinical systems were still in early development.
What was reported
Researchers from Shenyang University of Chemical Engineering and the Chinese Academy of Sciences' State Key Laboratory of Robotics designed an AR-based navigation system for minimally invasive spine surgery. The system integrates an O-arm CT scanner, a Polaris Vega optical tracker, and a Microsoft HoloLens headset. Preoperatively, CT images are reconstructed into a 3D spine model using thresholding and triangulation, then imported into Unity-3D for visualization and segmentation via PointNet++.
During surgery, the Polaris Vega tracker monitors reflective markers on instruments and the HoloLens, establishing coordinate transformations through a hand-eye calibration method adapted from robotics. The system uses a custom calibration tool with at least three spheres, maintaining distances over 50 mm for accuracy. Data is transmitted via TCP to a control console, which computes poses and sends them to the HoloLens for real-time overlay of the 3D lesion model and instrument position.
In spine model experiments, the system achieved a navigation error of less than 2.8 mm, meeting clinical requirements for spine surgery. The AR display provides an intuitive "see-through" view, helping surgeons locate lesions and guide instruments without continuous X-ray exposure.
Why it mattered
This work demonstrated a practical AR navigation approach that reduces reliance on 2D imaging and radiation, potentially improving surgical accuracy and safety. It also highlighted the transfer of robotics calibration techniques to medical AR, paving the way for broader adoption of AR in surgical navigation and other precision industrial tasks.
The experimental test on the spine model show that the system navigation error is less than 2.8 mm, which can meet the requirements of clinical application in spine surgery.
Source: 《机器人》期刊 (robot.sia.cn) · Published 2023-09-16 · “基于增强现实的脊柱微创手术导航系统”