The Research Of Helical Phase-Contrast CT Algorithm Base On Grating Imaging

Phase-contrast imaging is a state-of-the-art non-destructive detecting technology, it can provide high sensitivity, with which the internal structure of light elements can be observed, because the change of phase-shift is 1000 to 100000 times more than that of absorbing attenuation in hard X-ray wave band. Until now, the scientists have developed four Phase-contrast imaging methods, they are interferometer method, in-line method, diffraction enhanced imaging method and grating imaging method. The first three methods can not be used in the application area because of their own flaws, but grating imaging based on the Law of Lau can successfully use the incoherence source, which brought us into a new epoch of incoherence source phase contrast imaging. It has the potential to be utilized in many fields such as medicine, biology and material science etc.The conventional phase-contrast information extraction method, which is known as "phase stepping", is too complex, it needs to move the analyzer grating at least five times to expose in one projection angle, so this method will cost much longer time in the projection-acquisition process, which make grating imaging can hardly be used for human detection. The propose of "one-time exposure in one projection angle" solves the problem perfectly by decreasing the expose dose and simplifying the experimental complexity. In this thesis, by analyzing the feature of equivalence principle between parallel data and fan-beam data, a new phase-contrast CT method which is based on "one-time exposure in one projection angle" is developed by introducing the helical scanning into the field of phase-contrast imaging. First, every conjugation X-ray in the helical scanning is calculated and all the X-ray is put in the same plane by interpolation, then according to the property that the absorption of the conjugation X-ray is the same while the refraction angle of it is contrary, the phase-contrast information is extracted. At last the phase-contrast part is reconstructed by using filtered-back projection method. Computer simulations and performance analysis demonstrate the efficacy of the proposed method.