A RotBoost Generalized Multiclass SVM Classifier for Automated Face Shape Prediction in Orthognathic Plastic and Reconstructive Surgery During Computer-Assisted Diagnosis and Planning

Artificial intelligence-related researches aids in the planning and simulating orthognathic plastic and reconstructive surgery. It improves patient satisfaction, which is closely related to expectations. Plastic surgeons did not have an accurate solution for predicting the outcomes of post-operative shapes of patients before the surgery using their pre-operative face shapes. Hence, orthognathic plastic and reconstructive surgery requires an efficient framework for Computer-assisted planning to discover the correct patterns from the high volume of numerical and visuals clinical data that is available with respect to face shapes. Hence forth, a novel RotBoost generalized multiclass SVMs classifier involving a large dataset using the 3D morphable model is proposed in this paper. This proposed model assists in recognizing the orthognathic surgery patients' face from large data classified with the minimum false rates. Thus, high-quality 3D scans automatically classify orthognathic patients' faces and help assess pre-operative and postoperative face shapes. This process also assists the patients to predict the after shapes during medical planning even before commencing the surgery. The proposed approach provides superior performance compared with the existing techniques in accuracy, sensitivity, and specificity.