Condition monitoring based on sound feature extraction during bone drilling process

In order to enhance operation safety, this study analyzes the sound pressure signal obtained from the bone drilling process for condition monitoring. The drilling sound is caused by the contact between the drill bit and the bone, and we only pay attention to the harmonic components of the sound pressure signal whose frequencies are integer multiples of spindle frequency. Lifting wavelet transform is performed to extract the crucial characteristic from the sound pressure signal to correlate drilling condition. The product of standard deviation of wavelet coefficients is calculated to determine whether cortical or cancellous bone is drilled. The proposed method is experimentally verified through the drilling operation in in vitro porcine spines, and the results confirm the feasibility and achievable performance of the method and then the safety of the drilling operation is improved.