The objective of this research is to develop and evaluate 3-D quantitative microwave inverse scattering algorithms for dynamic dielectric environments, such as occur with use of a contrast agent in breast cancer screening, hormonal agents in cancer prevention, or various therapies in cancer treatment. The approach is to combine scattered field data from multiple scans over time of a dielectric environment into a single imaging problem through the use of spatial and temporal basis functions. Sparsity-inducing regularizers are applied to the basis coefficients to obtain high-resolution images of the changes in dielectric properties. The algorithms are validated using computational studies with 3-D MRI-derived numerical phantoms having a high level of anatomical realism and experimental studies with heterogeneous anthropomorphic physical phantoms.
Intellectual Merit: Innovative microwave inverse scattering algorithms are developed that exploit prior knowledge of the dynamic dielectric environment to overcome the limitations of the single-scan inverse scattering problem. Demonstration of these novel imaging algorithms using numerical and experimental testbeds provides evidence of their effectiveness.
Broader Impacts: This project has a positive impact on the health of women worldwide through development of improved breast imaging methods. The algorithms are also applicable to nondestructive evaluation. Engineering-grand-challenges case studies on breast cancer detection are developed for middle school and first-year undergraduate students. The project's emphasis on women's health is leveraged to recruit female students. Institutional programs are employed to involve minorities and undergraduates in the project. All students involved in the project receive interdisciplinary training at the interface of signal processing, electromagnetics, and medicine.
