WE‐A‐134‐01: Integration of X‐Ray and Optical Tomography for Precision Radiation Guidance and Real‐Time Treatment Evaluation

Purpose: We are integrating cone beam CT (CBCT) and bioluminescence tomography (BLT) onto a small animal radiation research platform (SARRP) to guide focal irradiation and online evaluate tumor response. Methods: X‐ray source provides up to 225kVp X‐ray for both CBCT imaging and radiation beam delivery. Imaging is performed in the order of CBCT, Diffuse Optical Tomography (DOT) and BLT. The CBCT provides anatomical information and DOT reconstructs the tissue optical properties, both of which serve as priori information for the BLT reconstruction. The CBCT system employs 65–100kVp x‐ray, a rotating animal stage and a CMOS detector panel with 200um pixel pitch enabling image resolution of 100um at the isocenter. A low‐noise CCD camera with an f/1.4 lens is used to capture light for optical imaging. DOT is conducted with a halogen lamp providing illumination through 9 fibers placed against the anterior surface of prone positioned animals. A novel rotating mirror system reflects light emitted from the animal to the stationary CCD camera enabling multi‐view acquisition. Four band‐pass filters with the selection of central wavelengths from 590nm–650nm are mounted in front of the lens allowing multi‐spectral BLT reconstruction. Results: The mouse CBCT showed clear differentiation of bone, lung, muscle and adipose tissue. BLT studies using tissue‐like phantoms indicate that the center of mass of an optical source can be localized to within 1 mm accuracy and source power can be restored with <10% error, even in the case of using only one single view/projection. The precision of image guidance will be validated in both phantoms and animals. Conclusion: The SARRP integrated with x‐ray and optical tomography provides accurate and precise image guidance for focal irradiation. With the capability of quantitative imaging, the system can be used to monitor tumor growth and treatment response online. The research is supported by the NCI grant R01CA158100‐01.