| Human surgery is often performed "blindly," without real-time assessment of diseased tissue that needs to be resected or normal tissue that needs to be avoided. However, recent progress in optical imaging suggests that image-guided surgery should be possible. Near-infrared (NIR) diffuse optical imaging takes advantage of deep photon penetration of NIR light into living tissue to provide fast, quantitative, relatively inexpensive imaging of endogenous and/or exogenous contrast at depth < 1 cm. In this dissertation, we explore the key parameters necessary to efficiently translate wide-field (typically > 5 cm in diameter) NIR optical imaging technologies to surgical environments.;Initial work focused on the translation of qualitative imaging of continuous-wave NIR fluorescence. For this purpose, a novel modular LED light source was designed to achieve high fluence rates over a 15 cm diameter field of view. The light source was integrated into a new, clinically friendly imaging system termed FLARE(TM) (fluorescence-Assisted Resection and Exploration), which provided one color, and two NIR fluorescence imaging channels, in real-time. FLARE(TM) was successfully translated to the clinic where it was evaluated for sentinel lymph node mapping during breast cancer surgery and breast reconstruction after mastectomy.;Motion artifacts are a considerable source of noise for both qualitative display of images and quantitative processing of image data. To solve this important problem, a novel, hardware-only gating device was developed. The device permits synchronized image acquisition from both non-periodic signals (e.g., involuntary motion) and periodic physiological signals (e.g., ECG, plethysmograph), with high timing precision.;Finally, technologies have been advanced that benefit several quantitative, wide-field imaging techniques for image-guided surgery. In temporal frequency-domain imaging (TFDi), the use of high power LEDs was investigated for performing fluorescence lifetime imaging (FLi) using a custom-made LED module capable of DC to 35 MHz modulation. In collaboration with Dr. Tromberg's group at the University of California, Irvine, a profilometry-based correction method that compensates sample curvature was designed for optical imaging. This correction technique was applied to spatial frequency-domain imaging (SFDi), and it was demonstrated that tissue optical properties and oxygenation could be extracted with good accuracy, under clinically realistic conditions. |
