PINNOCHIO: Physics-Informed Neural Network for Coupled Hyperelastic Interface-Volume Simulation in Orthognathic Surgery
Researchers have developed PINNOCHIO, a new framework using Physics-Informed Neural Networks (PINNs) to simulate facial soft-tissue deformation for orthognathic surgery planning. This approach addresses the accuracy-efficiency trade-off seen in existing methods by decoupling interface movements from volumetric deformation, enabling stable training and biomechanical consistency without requiring full volumetric ground truth. Tested on a cohort of 40 patients, PINNOCHIO demonstrated superior accuracy and physical validity compared to current baselines, offering a significant speedup over traditional Finite Element Methods. AI
IMPACT Enhances surgical planning accuracy and efficiency by providing a faster, more reliable simulation tool for complex soft-tissue deformations.