Compressed sensing (CS) is a new signal acquisition and processing theory. It can decrease the signal sampling time and computation cost by reducing the required data for signal recovery while maintaining good image quality. CS theory has drawn a lot of attention and made great progress in medical imaging. This paper introduces and summarizes the recent studies on CT reconstruction based on compressed sensing,which includes the analysis of the methods of traditional statistical iterative algorithm combined with compressed sensing,the methods of prior image constrained compressed sensing and the development course of dictionary learning. This paper also discusses the prospects of the development of this field.