Application of artificial intelligence methods for the reconstruction of the geometric model of the body of the human mandible

The tendency of personalized medicine is to adapt the medical treatment to each patient. Bone fracture treatment is best if personalized implants are used. For their design, there is a need for parametric 3D bone models that are adapted to a specific patient. In this the paper, the 3D parametric model of the body of the human mandible, defined as a model of points was created using the method of the anatomical entities, with the application of multiple regression. Mathematical relations between specific parameters (dimensions which are read on medical imaging) and anatomical orientations (measured values of coordinates of points) were established. In order to obtain greater accuracy in parameter prediction, this paper presents the application of the artificial intelligence method - neural networks for creating the 3D parametric model of the human mandible. The geometric accuracy of the created parametric models was tested using the deviation analysis in CAD software CATIA. The deviation analysis was performed between the initial and the resulting model. The results of the analysis are more than satisfactory.