Abstract

Integration of Multidimensional Functional Imaging with Mammography and MRI

In this project, we seek to understand better what combinations of structural, spectral, and dynamical information can best be employed to improve the ability of diffuse optical tomography (DOT) to diagnose breast cancer. We seek to do this combining DOT in a parallel plate geometry with other imaging modalities such as MRI and X-ray Mammography and with the spectral and temporal information that has become available with our next-generation DOT imaging systems. MRI and X-ray mammography are important first because the can proved a wealth of structural guidance to our DOT image reconstructions and second because they allow us to compare our DOT results with the co-registered simultaneous conventional measurements. This will allow us to quantify the improvements due to the different DOT imaging strategies we will consider. Spectral characterization is also important because our optical diagnostic information is ultimately derived from the wavelength dependence of the absorption and scattering of the different tissue chromophores (predominantly hemoglobin), which are expected to vary with disease. Finally, the temporal variations of tissue optical properties, brought about by either exogenous optical contrast agents or by the physiological manipulation of endogeneous tissue properties offers qualitatively new dimensions for tissue contrast and diagnostic interpretation.

Our effort is organized along three target areas. The first target area develops algorithms for image reconstruction that will be pursued in collaboration with the Software/Computation Core. Incorporation of spatial and spectral information into the reconstructions is an essential development for successful multi-modal imaging. The second target area focuses on the measurement of static breast optical properties. These measurements will be similar to conventional radiological breast imaging and we will use the results to optimize our imaging procedures and imaging hardware. These technologies will be advanced in collaboration with the Instrumentation core. The third target area explores our ability to collect and use additional dynamical (temporal) information. The additional contrast available from dynamical imaging will allow us to use more robust difference imaging techniques instead of single-frame imaging as is often done in breast imaging.

UPenn MRI/NIR platform for breast examination. Both breast are soft-compressed for double-sided MRI scan. Only the breast bearing the suspicious mass is optically scanned in transmittance geometry.

(B)
The MGH breast image. The bottom two units are our CW imagers (only the first 32 fibers are connected in this photograph). Top-left in a 19" rack are the two optical multiplexers and the RF imager. Top-right are the computers for data acquisition and processing.