Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ucin1233614556.pdf (2.24 MB)

ETD Abstract Container

Abstract Header

Deformation Analysis of Soft Tissues by Digital Image Correlation

Nageswaran, Ashok R.
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1233614556

Abstract Details

2009, MS, University of Cincinnati, Engineering : Mechanical Engineering.
Digital image correlation techniques are commonly used to evaluate deformation gradients from image sets by cross correlating pixel intensities. Recently, Bayesian framework based texture correlation technique have gained greater acceptance to study soft tissue deformation due to its ability to model noisy image data. Besides cross correlation, this probability based technique weighs regions of potential matches by radial distance from center of search region thereby eliminating large improbable displacements. Simple tension tests on soft tissue using magnetic resonance and ultrasound imaging modalities are presented. This initial study revealed encouraging results from ultrasound images compared to MR due to better a priori incremental image data. Having established a suitable imaging modality, plane strain tension and indentation tests were carried out on homogeneous (breast phantom) and non-homogeneous (muscle tissue) specimens and validated using optical images with surface markers. The specimens were deformed up to 10 mm in tension and indentation modes. The root mean square (RMS) error between the strain values from texture correlation and optical images varied from 6 to 10% for breast phantom compared to 2 to 7% for muscle tissue. As expected, the errors appeared to increase with larger deformation to suggest the limited scope of dependable estimates. The displacement estimates evaluated for square block sizes in muscle tissue, revealed acceptable RMS error range at 6 to 7 mm for tension and indentation test. The block sizes are significantly lower for homogeneous specimen due to better texture and dynamic range compared to muscle tissue. The indentation tests show the Bayesian texture correlation technique's ability to predict regions of non-homogenous deformation. Further studies must aim at analyzing image data with varying signal to noise ratios and identifying a measure of filter to remove questionable displacements.
Balakrishna Haridas, PhD (Committee Chair)
Liu Yijun, PhD (Advisor)
Douglas T. Mast, PhD (Committee Member)
Dong Qian, PhD (Committee Member)
67 p.
Biomedical Research
Digital Image Correlation; Bayesian probability; ultrasound;

Recommended Citations

Citations

  • Nageswaran, A. R. (2009). Deformation Analysis of Soft Tissues by Digital Image Correlation [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1233614556

    APA Style (7th edition)

  • Nageswaran, Ashok. Deformation Analysis of Soft Tissues by Digital Image Correlation. 2009. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1233614556.

    MLA Style (8th edition)

  • Nageswaran, Ashok. "Deformation Analysis of Soft Tissues by Digital Image Correlation." Master's thesis, University of Cincinnati, 2009. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1233614556

    Chicago Manual of Style (17th edition)

ucin1233614556
1,052
© 2009, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.