A Small Animal Optical Tomographic Imaging System with Omni-Directional, Non-Contact, Angular-Resolved Fluorescence Measurement Capabilities

The overall goal of this thesis is to develop a new non-contact, whole-body, fluorescence molecular tomography system for small animal imaging. Over the past decade, small animal in vivo imaging has led to a better understanding of many human diseases and improved our ability to develop and test new drugs and medical compounds. Among various imaging modalities, optical imaging techniques have emerged as important tools. In particular, fluorescence and bioluminescence imaging systems have opened new ways for visualizing many molecular pathways inside living animals including gene expression and protein functions. While substantial progress has been made in available prototype and commercial optical imaging systems, there still exist areas for further improvement in the outcome of existing instrumentations. Currently, most small animal optical imaging systems rely on 2D planar imaging that provides limited ability to accurately locate lesions deep inside an animal. Furthermore, most existing tomographic imaging systems use a diffusion model of light propagation, which is of limited accuracy. While more accurate models using the equation of radiative transfer have become available, they have not been widely applied to small animal imaging yet. To overcome the limitations of existing optical small animal imaging systems, a novel imaging system that makes use of the latest hardware and software advances in the field was developed. At the heart of the system is a new double-conical-mirror-based imaging head that enables a single fixed position camera to capture multi-directional views simultaneously. Therefore, the imaging head provides 360-degree measurement data from an entire animal surface in one step. Another benefit provided by this design is the substantial reduction of multiple back-reflections between the animal and mirror surfaces. These back reflections are common in existing mirror-based imaging heads and tend to degrade the quality of raw measurement data. Furthermore, the conical-mirror design offers the capability to measure angular-resolved data from the animal surface. To make full use of this capability, a novel equation of radiative transfer-based ray-transfer operator was introduced to map the spatial and angular information of emitted light on the animal surface to the captured image data. As a result, more data points are involved into the image reconstructions, which leads to a higher image resolution. The performance of the imaging system was evaluated through numerical simulations, experiments using a well-defined tissue phantom, and live-animal studies. Finally, the double reflection mirror scheme presented in this dissertation can be cost-effectively employed with all camera-based imaging systems. The shapes and sizes of mirrors can be varied to accommodate imaging of other objects such as larger animals or human body parts, such as the breast, head, or feet.