Tactile sensing of shape : biomechanics of contact investigated using imaging and modeling

Abstract

The overall goal of this research effort is to improve the understanding of the biomechanics of skin as it pertains to human tactile sense. During touch, mechanoreceptors beneath the skin surface are mechanically loaded due to physical contact of the skin with an object and respond with a series of neural impulses. This neural population response is decoded by the central nervous system to result in tactile perception of properties such as the shape, surface texture and softness of the object. The particular approach taken in this research is to develop a realistic model of the human fingertip based on empirical measurements of in vivo geometric and material properties of skin layers, so that the mechanical response of the fingertip skin to different shapes of objects in contact can be investigated, to help identify the relevant mechanism that triggers the mechanoreceptors in tactile encoding of object shape. To obtain geometric data on the ridged skin surface and the layers underneath together with their deformation patterns, optical coherence tomography (OCT) was used to image human fingertips in vivo, free of load as well as when loaded with rigid indenters of different shapes.

(cont.) The images of undeformed and deformed finger pads were obtained, processed, and used for biomechanically validating the fingertip model. To obtain material properties of skin layers, axial strain imaging using high frequency ultrasound backscatter microscopy (UBM) was utilized in experiments on human fingertips in vivo to estimate the ratio of stiffnesses of the epidermis and dermis. By utilizing the data from OCT and UBM experiments, a multilayered three dimensional finite element model of the human fingertip composed of the ridged fingerpad skin surface as well as the papillary interface between the epidermis and dermis was developed. The model was used to simulate static indentation of the fingertip by rigid objects of different shapes and to compute stress and strain measures, such as strain energy density (SED), and maximum compressive or tensile strain (MCS, MTS), which have been previously proposed as the relevant stimuli that trigger mechanoreceptor response.

(cont.) The results showed that the intricate geometry of skin layers and inhomogeneous material properties around the locations of the SA-I and RA mechanoreceptors caused significant differences in the spatial distribution of candidate relevant stimuli, compared with other locations at the same depths or the predictions from previous homogeneous models of the fingertip. The distribution of the SED at the locations of SA-I mechanoreceptors and the distribution of MCS/MTS at the locations of RA mechanoreceptors under indentation of different object shapes were obtained to serve as predictions to be tested in future biomechanical and neurophysiological experiments.