Sound Pressure Signal-Based Bone Cutting Depth Control in Robotic Vertebral Lamina Milling

Abstract

This paper aims to improve the performance of the robot-assisted bone milling operation. Based on a standard spinal surgery procedure, we established a dynamic bone milling model, which indicates that the cutting depth can be monitored based on the amplitudes of some specific harmonics of the milling sound pressure signal. The nominal frequency of these harmonics is an integer multiple of the spindle rotation frequency of the surgical milling tool. Therefore, we accurately extracted the absolute amplitude and frequency of the harmonics with nominal frequencies. We used the total harmonic amplitude to monitor the cutting depth and decrease the harmonic amplitude estimation error according to the absolute frequency of the first harmonic. A spine surgical robot carried out the bone milling experiment on artificial bone materials to verify. The results show that we can monitor the bone cutting depth with an accuracy of ±0.1 mm within the depth [0.1–1.2 mm] based on the auditory sense of the robot. The proposed method can improve the accuracy when a spine surgical robot performs the bone milling operation and make robot-assisted milling operation more accessible to inexperienced doctors via feedbacking the bone cutting depth in real-time.