Objective To investigate the value of knowledge-based iterative modal reconstruction(IMR) technology for coronary computed tomography on 256-slice spiral CT with radiation dose reduced to 80%. Methods Retrospective analysis coronary artery data of 21cases patients with heart rate is less than 65bpm, all patients underwent CCTA with retrospective ECG tagging with cardiac dose right helical scan. The data acquisition time window centered at the 75% of the R-R cycle, including a phase tolerance of 4-5％. The images were divided into four groups according to the phase and reconstruction algorithm. Data acquisition at End-systolic phase with iDose4 (group A), data acquisition at End-systolic phase with IMR (group B); data acquisition at mid-diastolic phase with iDose4 (group C), data acquisition at mid-diastolic phase with IMR (group D).then, the image noise(SD), signal-to-noise ratio(SNR) and contrast-to-noise ratio(CNR) of ascending aortic root were measured. Objective measurements were compared with one way analysis of variance, and subjective assessments were compared with Kruskal-Wallis H test. Results A total of 21 cases of image reconstruction were completed in the Endsystolic phase and mid-diastolic phase. Image noise of group C were significantly higher than other groups [(33.0±8.9)HU, (22.4±4.1)HU, (89.6±53.9)HU and (27.7±6.1) HU]. Group C compare to other groups, the difference was statistical significance (P <0.01), SNR[(25.4±8.2), (21.4±4.7), t=-2.92, -10.85] and CNR[(37.4±13.3), (36.5±18.1), t=-2.37, -5.60] of proximal coronary arteries in IMR were significantly higher than iDose4 which used the same radiation dose, and there was no significant difference between the group A and group D in SNR[(19.0±6.0) vs.(21.4±4.7), t=- 1.48] and CNR[(29.5±7.7) vs.(36.5±18.1), t=-1.64](all P>0.05).there were significant differences for four groups [(3.95±0.22), 4.0,(19.5±0.67), (3.81±0.4), H=69.12, P<0.05]in mean score, and with no significant differences for group A, B and D(H=5.56, P>0.05). Conclusion IMR can significantly reduce image noise, increase SNR and CNR, and improve image quality. The image quality of coronary artery can still be maintained at iDose4 after reconstruction with a significant reduction in radiation dose.