In-vivo imaging of breast tissue viscoelasticity

Mallika Sridhar, Jean K. Tsou, Michael Insana

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Viscoelastic imaging techniques are being developed to measure mechanical features of breast tissues using standard ramp-and-hold stimuli for creep measurements. This paper discusses a 3-person volunteer study to detail experimental methods used to measure breast viscoelasticity. It also summarizes our recent clinical data with 25 breast mass patients. We use a linear array transducer mounted to a surface force sensor to apply a ramp-and-hold force to the breast surface while recording RF frames at 2 fps. In particular we determine (a) the linear range of breast tissue deformation, (b) our ability to apply and hold a constant force for up to 200 s with and without force feedback, and (c) characteristics of viscoelastic creep in breast tissue, all using a new scanning technique that allows for normal breathing with minimal artifacts. Linear responses were observed for breast up to 4% strain. Below 5N of applied force, the creep response displayed classical arrheodictic behavior of polymer solids, settling to a constant strain after 100 s. We model the creep response as the sum of many exponentials that can be parameterized by a low-order Voigt model to obtain the retardation times, T1=3.2±0.8 s and T 2=42.0±28 s for normal breast tissue. The magnitude of creep is equivalent to the initial elastic response and hence is not significantly affected by small systematic errors in the stimulus. Newer scan techniques therefore extend acquisition times an order of magnitude to increase the model order and the information content.

Original languageEnglish (US)
Title of host publication2006 IEEE International Ultrasonics Symposium, IUS
Number of pages4
StatePublished - 2006

ASJC Scopus subject areas

  • Acoustics and Ultrasonics


Dive into the research topics of 'In-vivo imaging of breast tissue viscoelasticity'. Together they form a unique fingerprint.

Cite this