Identification of the Mechanical Properties of Living Skin: An Instrumentation and Modelling Study

Show simple item record

dc.contributor.advisor Nielsen, P en
dc.contributor.advisor Taberner, A en
dc.contributor.advisor Nash, M en
dc.contributor.author Parker, Matthew en
dc.date.accessioned 2017-08-07T21:15:13Z en
dc.date.issued 2016 en
dc.identifier.uri http://hdl.handle.net/2292/34867 en
dc.description.abstract The characterisation of soft tissue mechanics has the potential to benefit patients across a variety of medical conditions and procedures. Mechanical models of soft tissues may be used to predict surgical outcomes, identify lesions, assess the efficacy of treatments, and to optimise patient-specific treatment strategies. Characterisation of soft tissues that benefits individual patients requires the collection of force and displacement data, which should be gathered non-invasively, in vivo. Interpretation of mechanical data requires numerical models, due to the complex nonlinear, viscoelastic, anisotropic, and heterogeneous behaviour of skin. In this thesis, experimental-modelling frameworks are presented, utilising both lumped-parameter, and finite-element, models. A six-axis force-displacement microrobot was modified to reliably apply dynamic forces (up to 200 Hz), in vivo, in indentation and extension experiments, while a 4-camera stereoscope was developed and validated to measure the resulting surface deformation field. A surface profiling algorithm was developed to identify the geometry of skin placed in an experimental setup, and to identify the position of the microrobot’s indenter tip. Lumped parameter models were used to characterise the dynamic force-displacement behaviour of glabrous (on the palm) and hairy (on the anterior forearm) skin of ten volunteers. Individual directions were characterised by a Wiener model, identified through stochastic system identification, with variance accounted for ranging 94 % to 97 %. The use of stochastic system identification techniques provided a rapid means of characterising skin properties. Parameters were identified from 5-second samples, and the whole test procedure for full-scale perturbations lasted under 2 minutes per direction. The within-subject coefficients of variation (CV) of the Wiener static nonlinearity parameters provide insight into the reliability of the device. With normal indentation, the microrobot produced CVs ranging between 2 % and 11 %. The performance for extension tests was less reliable, with CVs ranging from 2 % to 19 %. However, the CV within individuals under extension is still within the ranges reported for commercial devices, such as the Cutometer and Reviscometer. Linear dynamic models were also found using stochastic system identification, at incremental loads. Linear models reported a Young’s modulus of 63 kPa at small indentation depths and 460 kPa at greater depths on the forearm, and 170 kPa to 1090 kPa for the palm. These values suggest that, at small indentations, perturbations were mostly made to the more compliant superficial layers in the skin, such as the hypodermis, before the stiffer layers were progressively recruited, such as the dermis and underlying tissues, such as muscle. The microrobot and associated analytic techniques provide a unique system to mechanically analyse the nonlinear, anisotropic, viscoelastic, and heterogeneous properties of skin. It is the first device to employ stochastic system identification approaches in multiple directions without the need to reconfigure or reposition the probe relative to the skin. The results demonstrate its ability to measure skin properties in an efficient and reliable manner. Lumped parameter models are difficult to relate to the underlying structure of skin. Finite element (FE) models were developed, which utilised constitutive relationships, to recreate skin geometry and predict surface displacements resulting from applied forces. FE meshes were fit to surface geometry data recorded from the stereoscope. Force boundary conditions were applied to FE nodes that were in contact with the indenter. The workflow was validated using controlled phantom studies. A single layer silicone gel phantom, and a composite silicone gel/rubber membrane phantom, were indented to 2.8 mm and 2.1 mm, respectively, and modelled in a finite element modelling package. Neo-Hookean models were selected for each material. Predictions of the surface deformation field were generated and compared to stereoscopic measurements. A least-squares nonlinear optimisation procedure was implemented, which minimised the difference between the model predictions and stereoscopic measurements. The identified parameter for the single-layer and two layer gel phantoms lay between the ranges of parameters found by independent measurements, and reproduced the surface deformations with an RMSE of 143 μm and RMSE of 138 μm, respectively. These findings demonstrate that a composite model can accurately predict the behaviour of a thin skin layer, tightly coupled to a thick bulk layer. The FE modelling approach was applied to the forearm skin of a healthy volunteer, using a set of 1.5 mm displacement, in-plane and out-of-plane, experiments. A layered, 3D, quadratic Lagrange, FE mesh was used to model the experimental geometry. A Gasser-Ogden-Holzapfel constitutive model was used to calculate the surface deformations resulting from the application of boundary forces. A displacement-weighted mean-square error objective function was constructed for the deformation field. The FE model was able to recreate the nonlinear, anisotropic, viscoelastic, and heterogeneous behaviour with a RMSE of 211 μm. The use of stereoscopic data offered improved identifiability over traditional single-displacement measurement approaches, as demonstrated by Hessian identifiability metrics. However, the model was insufficient to capture skin’s mechanical behaviour over the full force-displacement curve. The model may be improved by adding anisotropic pre-stress, and/or using a different constitutive equation. Improved constitutive models may be used in this workflow to predict surgical outcomes, identify lesions, assess the efficacy of treatments, and to optimise patient-specific treatment strategies. en
dc.publisher ResearchSpace@Auckland en
dc.relation.ispartof PhD Thesis - University of Auckland en
dc.relation.isreferencedby UoA99265057714102091 en
dc.rights Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated. Previously published items are made available in accordance with the copyright policy of the publisher. en
dc.rights.uri https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm en
dc.rights.uri http://creativecommons.org/licenses/by-nc-sa/3.0/nz/ en
dc.title Identification of the Mechanical Properties of Living Skin: An Instrumentation and Modelling Study en
dc.type Thesis en
thesis.degree.discipline Bioengineering en
thesis.degree.grantor The University of Auckland en
thesis.degree.level Doctoral en
thesis.degree.name PhD en
dc.rights.holder Copyright: The author en
dc.rights.accessrights http://purl.org/eprint/accessRights/OpenAccess en
pubs.elements-id 645161 en
pubs.record-created-at-source-date 2017-08-08 en


Files in this item

Find Full text

This item appears in the following Collection(s)

Show simple item record

Share

Search ResearchSpace


Browse

Statistics