Contrast agents with distinctive absorption and emission spectra, in combination with multispectral near-IR imaging, may provide a mechanism for the detection of breast cancer. While there is evidence of preferential drug accumulation at a tumor site, an important question is the concentration required to allow discrimination through tissue. An estimate of agent absorption effects is obtained from the solution of the diffusion equation in homogeneous tissue. In this paper absorption signatures derived from the diffusion equation and Monte Carlo simulation of a near-IR contrast agent, indocyanine green, are compared. Tradeoff curves are generated among the key relevant parameters; contrast, depth, and agent concentration. It is also shown that the diffusion equation solution for a localized contrast agent leads to an algorithm to estimate tumor location and depth from near-IR images. The algorithm is applied to in-vitro IR measurements of a tissue sample with an injected contrast agent. The results have application to the design of contrast enhancing drugs and associated discrimination algorithms.
展开▼
