Autonomous penetration detection for bone cutting tool using demonstration-based learning

In orthopedic surgery, bone-cutting procedures are frequently performed. However, bone-cutting procedures are very risky in cases where vital organs or nerves exist beneath the target bones. In such cases, surgeons are required to determine the depth of the penetration into the bone by using only their haptic senses. Thus, we developed a handheld bone-cutting-tool system that detects the penetration of the cutting material. The developed system autonomously detects the penetration before total penetration and stops the actuation of the cutting tool, leaving a very thin remnant of work material. The developed system estimates the cutting resistance by using its motor's current and rotational speed. On the basis of data collected preoperatively, the system estimates the cutting state by using a support vector machine (SVM). According to the SVM outputs, the system detects the penetration of the work material and autonomously stops the actuation of the cutting tool. The proposed method was verified through experiments, and the results showed that the developed system successfully detected the penetrations of work materials and stopped autonomously immediately before total penetration. This study showed that the autonomous detection of bone penetration with a hand-held bone-cutting tool is feasible by using the proposed scheme.