Correlation between intraosseous thermal change and drilling impulse data during osteotomy within autonomous dental implant robotic system: An in vitro study
Ruifeng Zhao,Rui Xie,Nan Ren,Zhiwen Li,Shengrui Zhang,Yuchen Liu,Yu Dong,An-An Yin,Yimin Zhao,Shizhu Bai
DOI: https://doi.org/10.1111/clr.14222
Abstract:Objectives: This study aims at examining the correlation of intraosseous temperature change with drilling impulse data during osteotomy and establishing real-time temperature prediction models. Materials and methods: A combination of in vitro bovine rib model and Autonomous Dental Implant Robotic System (ADIR) was set up, in which intraosseous temperature and drilling impulse data were measured using an infrared camera and a six-axis force/torque sensor respectively. A total of 800 drills with different parameters (e.g., drill diameter, drill wear, drilling speed, and thickness of cortical bone) were experimented, along with an independent test set of 200 drills. Pearson correlation analysis was done for linear relationship. Four machining learning (ML) algorithms (e.g., support vector regression [SVR], ridge regression [RR], extreme gradient boosting [XGboost], and artificial neural network [ANN]) were run for building prediction models. Results: By incorporating different parameters, it was found that lower drilling speed, smaller drill diameter, more severe wear, and thicker cortical bone were associated with higher intraosseous temperature changes and longer time exposure and were accompanied with alterations in drilling impulse data. Pearson correlation analysis further identified highly linear correlation between drilling impulse data and thermal changes. Finally, four ML prediction models were established, among which XGboost model showed the best performance with the minimum error measurements in test set. Conclusion: The proof-of-concept study highlighted close correlation of drilling impulse data with intraosseous temperature change during osteotomy. The ML prediction models may inspire future improvement on prevention of thermal bone injury and intelligent design of robot-assisted implant surgery.