Mechanical characterisation of in vivo human skin using a 3D force-sensitive micro-robot and finite element analysis

Abstract

The complex mechanical properties of skin have been the subject of much study in recent years. Several experimental methods developed to measure the mechanical properties of skin in vivo, such as suction or torsion, are unable to measure skin's anisotropic characteristics. An experiment characterising the mechanical properties of in vivo human skin using a novel force-sensitive micro-robot is presented. The micro-robot applied in-plane deformations to the anterior forearm and the posterior upper arm. The behaviour of the skin in each area is highly nonlinear, anisotropic, and viscoelastic. The response of the upper arm skin is very dependent on the orientation of the arm. A finite element model consisting of an Ogden strain energy function and quasi-linear viscoelasticity was developed to simulate the experiments. An orthogonal initial stress field, representing the in vivo skin tension, was used as an additional model parameter. The model simulated the experiments accurately with an error-of-fit of 17.5% for the anterior lower forearm area, 6.5% for the anterior upper forearm and 9.3% for the posterior upper arm. The maximum in vivo tension in each area determined by the model was 6.2 Nm(-1) in the anterior lower forearm, 11.4 Nm(-1) in anterior upper forearm and 5.6 Nm(-1) in the posterior upper arm. The results also show that a finite element model with a neo-Hookean strain energy function cannot simulate the experiments with the same accuracy.

Multidirectional In Vivo Characterization of Skin Using Wiener Nonlinear Stochastic System Identification Techniques

“…4) align most closely with the long axis of the forearm. Flynn et al [28] demonstrated that although the proximal-distal axis of the arm produced the stiffest response, the perpendicular direction did not produce the least stiff response, which was found at 60 deg from the long axis. Likewise, we found the lowest modulus in direction 2, rather than the approximately perpendicular orientation of direction 3 relative to the long axis of the arm.…”

“…A three-axis microrobot, similar to that of Flynn et al [28] shown in Fig. 1, underwent modifications to its force transduction and timing hardware so that stochastic system identification techniques could be implemented to characterize the dynamic mechanical properties of human skin.…”

“…Dynamic system identification requires a deterministic control and measurement environment. Modifications to the device described by Flynn et al [9,28,29] included moving the timing control from the Microsoft Windows environment to LABVIEW 2011 REALTIME software (National Instruments, Austin, TX). Microrobot control signals and measurements were made on a compactRIO FPGA-based system (cRIO-9022 and cRIO-9114, National Instruments) running LABVIEW 2011 REALTIME software.…”

“…Errors due to relocation can be avoided if the system does not require reconfiguration of the testing device. A microrobotic device capable of performing both normal and tangential deformations has been described by Flynn et al [28]. Similar to the device developed by Chen and Hunter, dynamic force inputs are provided via voice-coil actuators.…”

Non‐invasive evaluation of skin tension lines with elastic waves

“…However, the possibilities for in vivo testing are limited by technical feasibility oand by ethical considerations. When attempts are made at characterising human skin in vivo through mechanical means (stretching (13,14), suction (15,16,17), torsion (18,19), indentation (20), etc. ), the skin is distorted, rotated and/or stretched, which briefly deforms the collagen fibre network and thus, significantly affects the accuracy of the experimental measurements.…”