The common technique for extracting valuable substances in plants is the use of organic solvents such as Soxhlet which is very time-consuming and consumes a large amount of solvent and damages thermo-sensitive compounds due to high temperature. But for medicinal applictions, a suitable method for extraction without utilization of toxic organic solvents is needed. Today, supercritical extraction technology is a suitable alternative to traditional extraction methods that extract biologically active compounds from plants. Supercritical extraction, in particular extraction with carbon dioxide, has been considered in recent decades due to its advantages such as non-flammability, non-toxicity, high purity, low prices, lack of solvent residue, higher selectivity, higher speed, and higher yields. Scopoletin is a phenolic coumarin compound that is one of the main components of leaves of Althaea Officinalis and has properties such as anti-tumor, anti-thyroid, anti-diabetic, anti-inflammatory and pain, anti-fungal, Anti-convulsant, and strong antioxidant properties. Therefore, the necessity of extracting scopoletin is determined. In this study, extraction of Scopoletin from leaves of Althaea Officinalis using supercritical carbon dioxide and ethanol as co-solvent was performed on a laboratory scale and subsequently the results were compared with Soxhlet method as the base case of 100% recovery. The extracted samples were analysed by High-performance liquid chromatography (HPLC) and The response surface method was used to analyze the results of extraction. For this purpose, the effect of parameters such as pressure (14-30 MPa), temperature (30-70 °C), flow rate (0.9-2.1 ml/min), and dynamic time (50-130 min) on the rate Recovery of Scopoletin was investigated. The response surface analysis showed that the laboratory data was well fitted with a second-order polynomial model, with a determination coefficient of 95.92% and a correction coefficient of 92.35%. In addition, it was found that in the proposed model the linear term of flow rate, linear and second order terms of pressure, temperature and dynamic time and the interaction between temperature and pressure, have a very important effect on the recovery rate of Scopoletin and second order of flow rate has relatively high importance. The other interactions between the parameters are not significant. Also, the investigations showed that the optimal values of extraction parameters in the experimental range were, temperature 45.75 °C, pressure 22.88 MPa, flow rate 1.72 ml/min and dynamic time 100.9 min, that under these conditions, recovery 46.57% was achieved. Experimental results, with three replications under these optimal operating conditions, showed an efficiency of 44.63%, which is close to the modeling results. Finally, the process was experimentally modeled using an artificial neural network method. In the artificial network model, the optimal number of hidden layer neurons was determined to be 6 and the BR training algorithm was introduced as the best-performing algorithm so that its determination coefficient (R 2 ) was equal 99/44% achieved. Comparing the results of determination coefficient of response surface and artificial neural network, it is concluded that artificial neural network method performs better.