The invention relates to an
oral implant positioning method and
system based on
deep learning. According to the method, three-dimensional
point cloud data of an
oral cavity is obtained and preprocessed, and a pre-trained three-dimensional
deep learning segmentation model is utilized to automatically and accurately segment tooth, gingiva and alveolar
bone structures. And based on the segmentation result, extracting key form and
spatial relationship characteristics of the alveolar bone, and matching with an
implant database to primarily select a candidate scheme. By integrating geometric constraint
verification and biomechanical
simulation, the candidate schemes are optimized and decided, and the optimal
implant implantation position and posture meeting the safety and biomechanical requirements are calculated. And finally, automatically generating digital operation guide plate data completely matched with the gingival form of the patient according to the optimal scheme. According to the method,
automation and intellectualization of the whole process from data collection to surgical guide plate generation are achieved, the defects that a traditional method depends on artificial experience, is high in subjectivity and low in efficiency are effectively overcome, and the precision, efficiency and reliability of implantation
surgical planning are remarkably improved.