Geometrical co-calibration of a tomographic optical system with CT for intrinsically co-registered imaging.

摘要

A mathematical approach for geometric co-calibration of a dual-modal small-animal imaging system is presented. The system comprises an optical imaging setup for in vivo bioluminescence and fluorescence detection, as well as an x-ray CT, both mounted on a common rotatable gantry enabling fully simultaneous imaging at axially overlapping fields-of-view. Geometric co-calibration is performed once by imaging a single cylindrical light-emitting source with both modalities over 360 degrees at two axial positions, respectively. Given the three-dimensional coordinates of the source positions in the reconstructed CT volume data along with their two-dimensional locations projected at the optical detector plane, the following intrinsic system parameters are calculated: (i) the intrinsic geometric parameters of the optical detection system-five parameters for each view and (ii) the relative positional relationship between the optical and CT systems-two parameters for each view. After co-calibration is performed, experimental studies using phantoms demonstrate the high degree of intrinsic positional accuracy between the optical and CT measurements. The most important advantage of this approach is that dual-modal data fusion is accomplished without any post-registration strategies.