Augmented Reality Surgical Navigation in Minimally Invasive Spine Surgery: A Preclinical Study
Xin Huang, Xiaoguang Liu, Bin Zhu, Xiangyu Hou, Bao Hai, Dongfang Yu, Wenhao Zheng, Ranyang Li, Junjun Pan, Youjie Yao, Zailin Dai, Haijun Zeng
Abstract
BACKGROUND: In minimally invasive spine surgery (MISS), where the surgeon cannot directly see the patient's internal anatomical structure, the implementation of augmented reality (AR) technology may solve this problem. METHODS: We combined AR, artificial intelligence, and optical tracking to enhance the augmented reality minimally invasive spine surgery (AR-MISS) system. The system has three functions: AR radiograph superimposition, AR real-time puncture needle tracking, and AR intraoperative navigation. The three functions of the system were evaluated through beagle animal experiments. RESULTS: The AR radiographs were successfully superimposed on the real intraoperative videos. The anteroposterior (AP) and lateral errors of superimposed AR radiographs were 0.74 ± 0.21 mm and 1.13 ± 0.40 mm, respectively. The puncture needles could be tracked by the AR-MISS system in real time. The AP and lateral errors of the real-time AR needle tracking were 1.26 ± 0.20 mm and 1.22 ± 0.25 mm, respectively. With the help of AR radiographs and AR puncture needles, the puncture procedure could be guided visually by the system in real-time. The anteroposterior and lateral errors of AR-guided puncture were 2.47 ± 0.86 mm and 2.85 ± 1.17 mm, respectively. CONCLUSIONS: The results indicate that the AR-MISS system is accurate and applicable.