Three-dimensional virtual planning in mandibular advancement surgery: Soft tissue prediction based on deep learning

The study aimed at developing a deep-learning (DL)-based algorithm to predict the virtual soft tissue profile after mandibular advancement surgery, and compare its accuracy with mass tensor model (MTM). Subjects who underwent surgery were enrolled divided into training group test group. DL was trained using 3D photographs CBCT data based on surgically achieved displacements (training group). Soft simulations generated by MTM actual surgical jaw movements (test group) compared soft-tissue profiles postoperative distance mapping in terms of mean absolute error lower face, lip, chin regions. 133 subjects included — 119 14 for DL-based face region 1.0 ± 0.6 mm significantly (p = 0.02) MTM-based (1.5 0.5 mm). can following surgery. With clinically acceptable error. Therefore, it seems be relevant option prediction orthognathic options.