SECO Program Solicitation NSF 13-551: Microwave Nearfield Radar Imaging (NRI) Using Digital Breast Tomosynthesis (DBT) for Non-Invasive Breast Cancer Detection.

Hxi, Llc, Harvard MA

Investigators

Abstract

Early, accurate detection of breast cancer has long been a top medical priority. Recent clinical success using Digital Breast Tomosynthesis (DBT) developed by our Massachusetts General Hospital (MGH) research partners shows reduced false alarms relative to conventional mammography. However, despite DBT?s enhanced detection capability, the 1% radiological contrast between cancerous tissue and commonly-occurring fibroglandular tissue limits the unequivocal characterization of diseased tissue. Alternatively, due to the relatively high contrast (10%) between cancer and fibro-glandular tissue, microwave imaging has a strong potential to distinguish diseased from healthy tissue. However, the often highly heterogeneous nature of breast tissue, where fibroglandular tissue is randomly interspersed in the adipose background, results in disorganized microwave images resulting in degraded image reconstruction. In collaboration with our research partners at Northeastern University?s Gordon Center for Subsurface Sensing and Imaging Systems (Gordon-CenSSIS) ERC, HXI proposes to fuse the DBT X-ray imagery with microwave Nearfield Radar Imagery (NRI), and by building upon the image extraction techniques developed by ERC platform technology, to significantly improve the cancerous tissue detection rate. The proposed activities will help the scientific/medical community to advance the knowledge and understanding of the effectiveness of microwave NRI method in detecting breast cancer. Our approach is based on the use of high resolution X-ray-based DBT imaging to obtain the spatial organization of heterogeneous breast tissues and on coupling it as prior information to NRI for 3-D reconstruction. This completely new transformative concept has not been previously investigated, and it has the potential to overcome the limitations of a single-modality breast cancer imaging technology. This project brings together expertise from multiple disciplines: 1) HXI, world experts in radar design, will build the NRI sensor and antenna array; 2) CenSSIS will provide its platform technology algorithms to perform the sensing, imaging and feature extraction of biological tissues and conduct phantom experiments; and 3) Consultants from the MGH will provide medical expertise. Our approach, using new but validated imaging systems, allows for rapid transition to industrial product development. The project furthers the ERC?s strategic plan through the application of translational research for commercialization, with appropriate licensing of CenSSIS? IP to HXI. Additionally, the NRI/DBT system has the potential to expand HXI?s portfolio of marketable microwave and RF products. The broader impacts of this project are multifaceted, including improving breast cancer detection rates, minimizing the cost of treatment by reducing the number of false-positive screenings and preventing patients from going through the anxiety that callback examinations create. Today, mammography allows 10-15% of cancers to go undetected and present themselves within one year. In the United States approximately 45 million mammograms are performed annually with false positive rates of 8-10%. This equates to more than 4 million callback examinations per year. With the global market growing at a CAGR of 15.4% developing better imaging systems is a high priority, in terms of women?s health, emotional toll, and medical costs.