AI-Guided 3D Printing Revolutionizes Surgical Preparation

Researchers at Washington State University (WSU) have introduced a groundbreaking AI-driven 3D printing technique aimed at assisting surgeons in creating detailed replicas of human organs for pre-surgical practice. This innovation promises to enhance surgical outcomes by providing doctors with more precise tools for preparation, according to the NVIDIA Technical Blog.

AI Algorithm Enhances 3D Printing

The AI algorithm, which was trained on images and attributes of human kidneys and prostates, including weight, size, porosity, and vascular architecture, collaborates with 3D printers to optimize the printing process. It helps determine the best settings for accuracy, weight, and print speed, significantly improving the efficiency and precision of 3D printed models.

Kaiyan Qiu, an assistant professor of mechanical and materials engineering at WSU and co-author of the study, highlighted the time-saving potential of this technology. “For pre-surgical organ models, we know surgeons will need high fidelity models that can be printed out quickly and with low labor intensity,” Prof. Qiu explained. “We imagine a scenario where a surgeon receives an MRI and CT scan [of a patient] in the morning. She has two hours to prepare everything for surgery. The AI can optimize the parameters, and print out a model organ in half-an-hour, and the surgeon can then spend the remaining time practicing [on the organ replica].”

Optimization with Bayesian Methods

The research team, including WSU computer science professor Jana Doppa, employed a multi-objective Bayesian Optimization (BO) approach using BoTorch to enhance the 3D printing process. The BO algorithm utilizes a probabilistic surrogate model to approximate the relationship between printing parameters and the quality of the printed organ models, capturing uncertainties and enabling more robust optimization.

The AI model training was conducted using NVIDIA A40 GPUs, and NVIDIA NGP Instant NeRF was utilized to reconstruct a mesh object of the 3D printed model. This advanced AI process is not limited to medical applications; it can also be used to create prototypes for implantable medical devices, airplane and robot parts, batteries, and even customized shoes.

For more information on this research, visit the NVIDIA Technical Blog.