Artemisinin is a compound extracted from Artemisia annuaL. with a remarkable curative effect against malaria. It can be extracted using ultrasound-assisted extraction (UAE) and then detected via HPLC. In this study, response surface methodology (RSM) was used to optimize UAE conditions for obtaining the maximum yield of artemisinin. Three independent variables (ratio of solvent to material, extraction temperature, and ultrasonic power) were evaluated using the Box-Behnken experimental design, with the yield of artemisinin as a response variable. Experimental data were highly fitted to a mathematical-regression model using multiple linear regression (MLR). Based on response surface plots, the three independent variables exhibited interactive effects on the yield of artemisinin. The optimal extraction conditions were as follows: 42.71mL/g ratio of solvent to material, 41.86 degrees C extraction temperature and 120W ultrasonic power. The predicted yield of artemisinin by model was 0.7848%, whereas the actual yield in the extracts was 0.7826%+/- 0.0790% in adjusted optimal conditions, with a relative error of 0.28%. The results undoubtedly demonstrated that RSM could be used to explore the optimum conditions of artemisinin extraction.