The invention relates to the technical field of CT image reconstruction, and discloses a
cone beam CT scanning image rapid reconstruction
system and method based on
deep learning, and the
system comprises a dual-
ray generation module, a scanning module, a
deep learning reconstruction module, and a result output module. The dual-
ray generation module comprises two
ray generators, a beam outlet of one ray generator is provided with a 0.1 mm low
atomic number material aluminum to filter an X-ray ultra-low energy part, and the other beam outlet is provided with a 0.1 mm
high atomic number material
copper to attenuate the X-ray low energy part. According to the method, the dual-energy-spectrum projection data is obtained through the dual-ray generator, the tissue linear
attenuation coefficient is solved through fitting by taking PMMA and aluminum as base materials, background interference is corrected in combination with air reference data, the error of the reconstructed CT value in the
soft tissue area is smaller than or equal to + / -10 HU, the error of the reconstructed CT value in the
cortical bone area is smaller than or equal to + / -30 HU, and compared with traditional CBCT, the method is more accurate, and a reliable foundation is laid for preventing
power ratio calculation.