Investigation into the automatic drilling of cortical bones using ANFIS-PSO and sensitivity analysis

During the bone drilling process, in case of excessive force can cause breakage, crack initiation and severe damage to bone tissue. Additionally, if drilling temperature rises higher than the generally believed 47 °C, it can cause thermal necrosis phenomenon which leads cell death of the bone tissue. In this paper, using helix angle and point angle as most influential factors, the evolution of process force and temperature are studied at the same time. For this purpose, using a central composite experimental design and preparing the required drills, the experimental tests have been performed on fresh bovine bone. Also, the present study benefits from an adaptive neuro-fuzzy inference system (ANFIS) whose parameters are optimized using the particle swarm optimization (PSO) method in order to predict force and temperature. To model the process behavior, the results of performed experiments for training (based on 70% of the data) and testing (based on 30% of the data) of the system are utilized. The findings of this research show that the adaptive neuro-fuzzy inference system is capable of properly predicting the force and temperature. Overall, the average errors in estimating the values of force and temperature in the training and test sections are 1.249 and 3.818%, respectively. After modeling with ANFIS-PSO, the Sobol sensitivity analysis method is used to quantitatively measure the effect of each input parameter over the process temperature and force. The results of sensitivity analysis show that with increasing both helix and point angles, the temperature in bone drilling process is decreased while the force is increased. In addition, it is proved that the sensitivity of temperature and force values to changes in helix angle is slightly higher than changes in drill point angle.