An Integrated Approach of Online Instant Surface Reconstruction for Intraoperative Printing on Living Cranial Defects

Rapid reconstruction of a curved defect model within the allocated surgery time is a critical aspect of intraoperative bioprinting for cranial defect treatment. However, current data acquisition and defect reconstruction methods often involve manual intervention, which is impractical for real-time intraoperative printing due to their intricate and time-intensive nature. To tackle this challenge, this study introduces an online acquisition approach and a real-time automatic reconstruction algorithm for intraoperative defect data. Initially, a surface preprocessing scanning technique is proposed to address bone surface reflections and complex noise from blood and cranial tissues in live defective skulls, ensuring the acquisition of reliable defect data. Subsequently, an integrated reconstruction algorithm is developed for the obtained defect data. This algorithm seamlessly incorporates various techniques, including range filtering-based point cloud denoising, defect edge detection, patch creation utilizing the boundary-center method, and defect surface optimization based on the harmonic function. These components enable the algorithm to autonomously achieve immediate defect surface reconstruction. Finally, the effectiveness of the integrated algorithm is demonstrated through its application in surgical procedures on live rat cranial defects and human cranial defect models, showcasing its feasibility, superior accuracy, and practicality compared to software-based reconstruction methods. The algorithm is user-friendly, catering to both medical professionals and non-professionals, and fulfills the requirements of clinical intraoperative cranial treatment.