Expert Consensus on Data Annotation and Quality Control for Trajectory Planning of Thoracolumbar Pedicle Screw Placement

In recent years, pedicle screw placement robotic systems developed with the integration of artificial intelligence (AI) technologies have demonstrated great potential to improve the accuracy and efficiency of preoperative trajectory planning and to enhance surgical safety. Such AI-assisted planning systems can precisely calculate the entry point, end point, and trajectory of screws, thereby effectively avoiding injury to critical anatomical structures and assisting surgeons in developing optimal surgical strategies. However, the performance of the underlying algorithms and their clinical applicability largely depend on datasets with high-quality annotations. To standardize the construction of such datasets, the National Institutes for Food and Drug Control (NIFDC), together with relevant institutions, has formulated this expert consensus, which aims to provide standardized guidance for the development of pedicle screw trajectory planning datasets.