Predict Afferent Tactile Neural Signal for Artificial Nerve Based on Finite Element Human Hand Model

Abstract

This paper aims to investigate the biomechanical aspect of human hand tactile perception by using finite element method and build the artificial neural nerve which can be interfaced with human afferent nerve. A subject-specific digital human hand finite element model (FE-DHHM) was developed based on CT and MR images. The geometries of phalanges, carpal bones, wrist bones, ligaments, tendons, subcutaneous tissue, epidermis and dermis were all included. The material properties were derived from in-vivo and in-vitro experiment results which are available in the literature, the boundary and loading conditions which were kinematic motion data and muscle forces, were captured based on the specific subject. This FE-DHHM was validated against in-vivo test results of the same subject based on contact pressure and contact areas. The whole active touch procedure was performed and simulated, the strain energy density near the locations of mechanoreceptors including slowly adapting type 1 (SA-I) and rapidly adapting (RA) were extracted and then used as inputs into the transduction and neural-dynamics (Izhikevivh neuro model) sub-model to predict neural spike or somatosensory information. A prototype of ‘artificial nerve’ which can produce the action potential signal is presented. Therefore the FE-DHHM presented in this paper can make a detailed and quantitative evaluation into biomechanical and neurophysiological aspects of human hand tactile perception and manipulation. The results obtained in this paper can be applied to design of bionic or neuro-robotic hand in the near future.