[en] Objective: To study the optimal radiographic conditions for patients with knee arthroplasty (TKA) and the feasibility of clinical application. Methods: According to the different exposure modes, including automatic exposure control(AEC) or manual exposure mode (FIXED), the human body specimen was exposed at a certain kV and different mA or mAs. The entrance surface dose (ESD) and dose exposure index(DEI) were recorded, and the subjective image quality was assessed. The test group with optimum parameters obtained with statistical analysis were compared with control group in radiation dose and subjective evaluation. Results: In AEC group, ESD and kV showed a negative correlation (r = -0.973, P < 0.05). At the same mA, DEI was firstly decreased and then increased with kV increasing. The subjective image quality score was(1.96 ± 0.56). DEI and physician ratings showed negative correlation (r = -0.840, P < 0.05). In FIXED group, ESD increased with the increasing mAs (r = 0.845, P < 0.05). DEI and kV,mAs indicated a positive correlation (r = 0.845, P < 0.05). The subjective image quality scores was (4.33 ± 0.79) without correlation with DEI. Compared with the control group, the subjective image quality scores of test group increased, and the radiation dose decreased significantly. Conclusion: It is optimal using manual exposure mode with 80 kV/1.25-1 mAs, 75 kV/2-1 mAs, 70 kV/2.5-1.25 mAs, 65 kV/3.2-1.6 mAs and 60 kV/3.2-2 mAs for patients with TKA. (authors)