Three-dimensional microwave breast imaging: dispersive dielectric properties estimation using patient-specific basis...

Area of Science:

• Medical Imaging
• Electromagnetics
• Computational Physics

Background:

• Breast imaging using microwave tomography requires estimating dielectric properties on a discrete mesh.
• High dimensionality in 3D imaging leads to significant computational challenges.

Purpose of the Study:

• To reduce the computational complexity of 3D breast microwave tomography.
• To enable efficient multi-frequency inverse scattering algorithms for breast imaging.

Main Methods:

• Replacing the discrete mesh with a small set of smooth basis functions.
• Estimating coefficients of basis functions instead of dielectric properties at each mesh element.
• Constructing basis functions using breast surface information to balance resolution and complexity.

Main Results:

• Achieved imaging accuracy comparable to the original discrete mesh in single-frequency microwave tomography.
• Demonstrated the feasibility of applying computationally efficient multi-frequency algorithms in 3D.
• Successfully verified the approach using two realistic 3D numerical breast phantoms.

Conclusions:

• The proposed basis function approach effectively reduces the dimension of the 3D microwave tomography problem.
• This method enhances computational efficiency for multi-frequency breast imaging.
• The technique offers a viable alternative for accurate and computationally tractable 3D breast imaging.