Prediction of anesthesia dosage, i.e., the propofol dosage is quite a complex task for anesthesiologists. Anesthesiologists need to consider many parameters in order to make a decision. İf any mistake arises in the process of making a decision, it leaves a long-term or short-term impact on the patient; in some cases, it may even endanger the life of the particular patient. To address this problem, deep learning models can be used to predict the dosage of propofol that needs to be given to a patient during surgeries to sedate them. This can be achieved with the help of already existing medical data recorded during the surgeries which consists of the parameters that the anesthesiologists generally use to make decisions and also the corresponding dosage given to the patient in that particular surgery. Five different deep learning models are trained to see which model can perfectly and accurately predict the propofol dosage. Among the five deep learning models, convolution neural network (CNN) showed better results with 0.925 for coefficient of determination (R2 value) and 0.001 for mean squared error (MSE). This study highlights the efficiency of deep learning models to forecast the decisions being made by anesthesiologists.