PROJECT SUMMARY
While dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) is recognized as the most sensitive
examination for breast cancer detection, it has a substantial false positive rate and injection of gadolinium (Gd)
contrast agents appears to have safety risks. Thus, alternatives to diagnosing breast cancer based on
endogenous contrast are of growing interest. The overall goal of this proposal is to develop and evaluate a
concurrent near-infrared spectroscopy (NIRS) and MRI imaging platform to explore whether the combined
modality, which does not require contrast injection or involve ionizing radiation, can achieve acceptable
diagnostic performance. The scientific premise of this proposal is underpinned by data from preliminary clinical
studies that show MR-guided (MRg) NIRS adds much needed diagnostic specificity to breast MRI. We will also
explore the exciting hypothesis that MRg NIRS serves as a surrogate for DCE MRI, and similar or the same
diagnostic performance can be obtained by adding MRg NIRS to non-contrast MRI sequences, thereby
eliminating the need for contrast injection. In preliminary studies, sensitivity, specificity, accuracy, and AUC of
non-contrast MRI when combined with T2-guided NIRS were 94%, 100%, 96%, and 0.95, respectively,
whereas these values were 94%, 63%, 88%, and 0.81 for DCE-MRI alone. This performance can be improved
even further through direct regularization from images (DRI) methods and multi objective priors we propose to
investigate in Aim 2. Equally important, we found the diagnostic performance of MRg NIRS is not degraded
when CW signals are used to form NIRS images as long as subject specific scattering estimates are obtained
(in Aim 1). Indeed, the practical utility of our current MRg NIRS system is undermined by its reliance on FD
data, and the long, large-core fiber bundles that transmit light to/from data acquisition hardware to the breast
inside the scanner. Elimination of these bulky weight-bearing fiber cables also simplifies the breast optical
imaging interface, and opens the door to development of new options for the innovative, clinical-friendly design
proposed in Aim 1. We will validate this platform iteratively in phantoms and volunteers (in Aim 3) and evaluate
the new MRg NIRS + MRI technology when combined with and without contrast injection in a clinical study of
60 patients with breast abnormalities. The project leverages an extensive infrastructure we have developed for
translational breast imaging where investigators from diagnostic radiology and biomedical engineering have
collaborated to develop and evaluate advanced technology for more than two decades.