Spectroscopic reconstructed near infrared tomographic imaging for breast cancer diagnosis.

摘要

Near-infrared (NIR) light has the potential to be used as a non-invasive means of diagnostic imaging within the human breast. Due to the diffusive nature of light in tissue, computational model-based methods are required for functional imaging within the breast. These methods can be used to reconstruct tomographic images of absorption and scattering properties of the interior of the breast from light measurements at the skin's surface. The work presented here describes improvements in both the data acquisition system and image reconstruction algorithm of NIR tomographic imaging for breast cancer diagnosis.;Specifically, the design, construction, and testing of a frequency-domain parallel-detection NIR data acquisition system is presented. Data acquisition from the system requires under 30 seconds for a single tomographic slice at one optical wavelength with a measurement repeatability for a single phantom on average of 0.5% in AC Intensity and 0.4 degrees in phase. In addition, methods for improved reconstruction in the areas of imaging additional tissue chromophores and the separation of absorption and scattering heterogeneity are described. The improved system and reconstruction methods have been tested extensively with computer simulation and tissue-simulating 'phantoms'. In addition, (forty-five) healthy volunteers, and (five) women with breast abnormalities have been imaged. Moderate-resolution quantitative tomographic images of absorption and scattering are reconstructed from data acquired at the breast tissue surface, and the quantitative absorption images are combined at multiple optical wavelengths to generate images of oxygenated and de-oxygenated hemoglobin, lipid, and water concentration. This research has led to an improved novel imaging system which is potentially useful in the diagnosis and detection of breast cancer.