Registration Workflows Accuracy For Robot-Assisted Osteotomy In Endodontic Microsurgery

This study evaluated the accuracy of robot-assisted osteotomy in endodontic microsurgery using different registration workflows. Twelve printed mandibular jaw models were divided into 4 group. U-shape tube (UT) groups is divided as UT-O and UT-B groups (O/B: registration points at occlusal or buccal site). Cone-beam computed tomography (CBCT) images of jaw models with UT were scanned. Customised template (CT) groups, divided as CT-O and CT-B groups. Each jaw information from CBCT scan and intraoral scan without any additional device were obtained. The osteotomy was performed with a robotic system. The postoperative CBCT images were obtained. Platform, apex, angular deviations and registration time were compared. Multilevel linear regression models were used to evaluate the accuracy outcomes of the registration devices (UT or CT), registration location (buccal site or occlusal site), tooth type (premolars or molars), and dentition site (right or left). The registration times were compared using the nonparametric test (P<0.05). A total of 72 sites were drilled in 12 models. Significant differences (P<0.01) among different registration location were found for all accuracy variables. No significant differences were found on other factors for all the accuracy variables. Less time was required for UT-B and CT-B groups. Robot-assisted osteotomy accuracy using UT registration workflow was similar to that of CT registration workflow. The buccal registration points were more accurate and efficient in robot-assisted osteotomy than occlusal registration points.