An inverse problem approach to stiffness mapping for early detection of breast cancer

作者：Olson Lorraine G^*; Throne Robert D

来源：Inverse Problems in Science and Engineering, 2013, 21(2): 314-338.

DOI：10.1080/17415977.2012.700710

• breast cancer
• genetic algorithm
• finite elements
• inverse problem
• elastography

Early detection of breast cancer will continue to be crucial in improving patient survival rates. Our ultimate goal is to develop a system that automates, quantifies and enhances the resolution of the manual breast exam. In this study, we examine computational techniques which use breast surface force and deflection measurements to create detailed maps of the elastic modulus of the interior of the breast tissue. This approach is a reformulation of our earlier two-dimensional technique to make it more practical and we extend the approach to three dimensions. Finite element methods were used to model the tissue response (reaction force and surface displacements) to applied surface indentations. A variety of test cases with assumed tumour locations were defined, and measured results' were created. Numerical noise was added to these simulated measurements at two noise levels. A genetic algorithm was developed to identify the distribution of tissue material properties within the breast given the measured' reaction forces and surface displacements. For our two- and three-dimensional model problems, tumours as small as 1cm could be detected reliably at a signal-to-noise ratio of 23dB.

• 出版日期2013-3-1