The invention discloses a multi-
parameter analysis-based maxillary first
molar extraction operation type decision-making method and a multi-
parameter analysis-based maxillary first
molar extraction operation type decision-making
system. The method comprises the steps of firstly obtaining
patient age and oral
panorama; secondly, segmenting the panoramic film through a trained DeepLabV3 + semantic segmentation model, and extracting a mesio-middle
cheek root outer side edge, a mesio-middle
cheek root inner side edge, a mesio-middle
cheek root outer side edge, a mesio-middle cheek root inner side edge and a root bifurcation point; calculating a
tooth root minimum curvature
radius R, a root bifurcation included angle theta and a local standardized
bone mineral density index D based on the extracted contour; and finally, entering a preset hierarchical decision engine according to the age of the patient, automatically outputting an operation form of suggesting overall extraction or root extraction in combination with the quantization parameter, and generating a decision basis and a quantization parameter report. According to the method,
deep learning feature extraction and a clinical
rule engine are combined, objective quantitative
decision making and personalized recommendation of a tooth extraction operation are realized, and the accuracy and
interpretability of
decision making are improved.