Algorithms and Systems for Electromagnetic and Ultrasound Inverse Problems

点击翻译按钮获取中文摘要

This NSERC Discovery Grant proposal aims to advance the core scientific knowledge and practical technologies that form the foundation of quantitative electromagnetic wave imaging (EWI) and ultrasound imaging (USI). Both are wavefield imaging techniques that create quantitative images of particular properties of an inaccessible object or region of interest. Properties such as the permittivity and conductivity of a region are imaged when one utilizes electromagnetic waves, e.g., microwaves, whereas ultrasonic properties, such as the compressibility, attenuation factor, and mass density, are imaged when using ultrasound waves. These property images can be utilized in a wide variety of diagnostic applications. Under my direction, the Electromagnetic Imaging Laboratory at the U. of Manitoba has made significant contributions to particular EWI and USI algorithms and we've built imaging systems applicable to various biomedical and industrial applications. The primary biomedical focus has been on building systems and technologies for the detection of breast cancer as well as for the frequent monitoring of breast cancer treatment. These same techniques have also been incorporated into stored-grain imaging systems that are used for the monitoring of grain quality so as to provide the early detection of spoilage. The quantitative wavefield imaging techniques we've advanced, based on solving the nonlinear inverse scattering problem, can be utilized in imaging applications wherein wave-type energy can be introduced as an interrogating field and the scattered-field outside the region can be accurately measured. Our system design research enables the practical manifestation of this interrogation/measurement process and our software development, which incorporates the imaging algorithms, allows us to produce the property images. Scientific and technological challenges limit the accuracy and resolution of images obtained using wavefield imaging. These challenges have limited the adoption of EWI for breast imaging but there is reason to believe that research will lead to improvements. Accuracy, rather than resolution, is more important for grain imaging where no competing technology that produces property images exists. The overall long-term focus of this research program is to improve and adapt EWI and USI algorithms, technologies, and systems to the point where these modalities become commercially viable modalities for both the breast imaging application as well as for the grain-imaging application. The main methodological theme is to simultaneously develop algorithmic advances while developing, incorporating, and utilizing system design features that aid the imaging process. The success of this research will open-up new biomedical and novel industrial applications (e.g., non-destructive evaluation). A total of seven graduate students will be involved in this research.