Quantifying machining outputs of pristine human teeth relevant to dental preparation procedures

Minimally-assisted tooth repair (MaTR) systems are envisioned to be capable of substituting for the skill of a dentist. If successfully developed, MaTR systems could enable lower-skilled dental technicians to provide dental care at a fraction of the overall medical cost. This paper explores a key initial step towards the development of such systems by quantifying the machining responses of pristine human teeth relevant to dental preparation procedures. The working hypothesis of the study is that such findings will enable the benchmarking of key process planning and control metrics relevant for the future development of MaTR systems. To this end, pristine human cadaver teeth were cut using a computer-controlled motion platform and dental hand-piece. Relevant cutting responses, such as cutting forces, in-process rotational speed of the dental bur, teeth morphology, and bur wear were captured. The trends in cutting forces show the potential for implementing region-specific process parameters for cutting the enamel and dentin regions of the tooth. A feed-per-tooth value of 0.1 µm at rotational speeds of 8 krpm and 50 krpm is seen to cut both the enamel and dentin regions at cutting forces lower than patient discomfort thresholds identified in literature. Cutting force signals were also successfully mapped against the CT-scan data of the tooth. This mapping indicates a clear identification of the enamel/dentin regions, and a transition region that is dependent on cutting parameters, tooth/tool geometry and tool pose. The trends in the in-process rotational speed of the dental bur indicate that stalling of the dental bur occurs at feed per tooth values greater than 0.25 µm. The evidence of stalling can be detected by both a drop in the cutting force signal and by surface morphology changes on the cut surface of the tooth. MaTR systems should be designed to avoid bur stalling regions by either operating at feed per tooth values ≤ 0.25 µm or by the use of dental spindles with higher torque capacity. Lastly, the type of fit present on the shank of the bur is seen to result in differences in the cutting force signals and wear of the cutting edges (flutes) of the dental bur. In general, a right-angle (RA) fit on the shank of the dental bur results in a larger tool runout leading to uneven loads on the flutes and increased tool wear. The friction grip (FG) fit avoids these problems and may be more suited for MaTR systems.