User login


You are here

Journal Club Forum for April 1st: Strain measurement in soft tissues

Many musculoskeletal soft tissues, such as tendons, ligaments, meniscus and cartilage are inhomogeneous. Hence, during mechanical loading it is likely that a nonuniform strain pattern occurs within the tissue. These nonuniform strain patterns may assist in successful load transmission and minimize rupture of the tissue during physiological loading. Determination of local material properties will likely be important for successful function and design of tissue engineered replacements. In the late 1980’s uniaxial tensile tests were conducted using a video camera in conjunction with surface markers to document local strain distributions on the surface of ligaments. Photoelasticity has also been used to document local strain patterns.

As Magnetic Resonance Imaging (MRI) technology has improved in the recent decade, its utility in determining local strains, noninvasively, within soft tissues has evolved. MRI will likely influence the biomechanics community with the capability to assess in vivo, 3-dimensional tissue strains.

One such MRI-based technique is DENSE-FSE or displacement encoding with stimulated echoes and a fast spin echo readout. The goal of this work was to acquire images with high spatial resolution in reasonable imaging times. This approach requires that the sample first be loaded to reach steady-state to avoid motion artifacts. Images are collected while the sample is cyclically loaded requiring a repeatable loading cycle to allow for sufficient time to collect the MR images. Hence, for a single, fast displacement test that is not repeatable, this approach may not work. Reducing the required resolution would shorten the imaging time.

Neu, C. P. and J. H. Walton (2008). "Displacement encoding for the measurement of artilage deformation." Magn Reson Med 59(1): 149-55.

Gilchrist et al., presents a texture correlation algorithm using first-order displacement mapping terms with MR images. This approach eliminates the need for "tags" or "markers". However, the technique is heavily dependent on image contrast/texture and thus is sensitive to noise.

Gilchrist, C. L., J. Q. Xia, et al. (2004). "High-resolution determination of soft tissue deformations using MRI and first-order texture correlation." IEEE Trans Med Imaging 23, 546-53.

Finite element warping has also been used to track local displacements on MR images. While this pproach eliminates the need for markers in the tissue, or MR tags, it does assume the discretized template to be a hyperelastic material in determining fiber stretch. Cine-MRI and deformable image registration uses differences in image intensities between a reference position image and a loaded position image to enerate a body force that deforms a FE representation of the template so that it matches the target. This requires a priori assumptions about material properties and constitutive behavior of the material and only works for static loading.

Phatak, N. S., Q. Sun, et al. (2007). "Noninvasive determination of ligament strain with deformable image registration." Ann Biomed Eng 35, 1175-87.

Major technical advances in MRI have allowed the above approaches to be developed for non-invasive measurement of soft tissue deformations. With continued advances in MRI technology that may increase resolution and shorten loading times, these approaches and others may enable dynamic three-dimensional collection of physiologically loaded soft tissues both in vivo and in vitro.

Zaoyang Guo's picture


Very interest topic. One of my research interests is to develop hyperelastic constitutive model for soft tissue. If the strain measurement requires the constitutive model to be pre-determined, then the experiment vs. model is like egg vs. hen. Which one should come first?

I cannot access the papers you mentioned currently, can you introduce more about the resolution of the strain measurement?

Many thanks,





MichelleLOyen's picture

H. Lu et al did some beautiful work on higher order effects in Digital Image Correlation, see

Deformation Measurements by Digital Image Correlation: Implementation of a Second-order Displacement Gradient

by H. Lu and P.D. Cary, Experimental Mechanics 40 (2000) 393.

Most algorithms for strain measurement are direct and do not require an assumed functional form to the constitutive model.

My concern is in the time-dependent mechanical behavior of soft tissues in this context.  If you deform the sample rapidly and use these data to  examine local strains, you might get a very different picture than if you examine the equilibrium configuration, where physical rearrangement of the tissue microstructure might be possible.  

Subscribe to Comments for "Journal Club Forum for April 1st: Strain measurement in soft tissues"

Recent comments

More comments


Subscribe to Syndicate