Low dose CT image statistical iterative reconstruction algorithms based on off-line dictionary sparse representation

Though reduced number of projection angles or lower current intensity of X-ray tube can reduce radiation dose and therefore relieve damage to human bodies, the former one will result in incomplete projection data while the later a declined signal to noise ratio of projection data. Two low-dose CT image reconstruction methods based on off-line dictionary sparse representation were proposed to solve the above problems. The offline dictionary was trained from the clear CT images which were selected from existing images, and saved for the low dose CT image reconstruction. For the limited angle scanned reconstruction, the iterative algorithm based on the trained dictionary was proposed to narrow the solution space and improve the reconstruction results. For the low X-ray tube scanned reconstruction, a statistical iterative reconstruction algorithm based on the trained dictionary was introduced to improve the performance of resisting noise. The experimental results show that both of the two approaches can precisely reconstruct CT images even when radiation dose is less than 10% of original amount.