Fig. 5From: A dosing strategy model of deep deterministic policy gradient algorithm for sepsis patientsFor differences in dosing combinations of IV fluids and VPs, each square represents the difference as a percentage of the total sample size. It can be seen that the difference in VP dosing is lower than that of IV fluids. It can be seen that the model is more inclined to make more drug dosing decisions. The model based on the DDPG algorithm recommended a sample size of 54.76% to receive more dosing (part a in Fig. 5), whereas the model based on the DQN algorithm recommended only 34.82% (part b in Fig. 5)Back to article page