Self-learning fuzzy control of atracurium-induced neuromuscular block during surgery.

Self-learning fuzzy logic control has the important property of accommodating uncertain, non-linear and time-varying process characteristics. This intelligent control scheme starts with no fuzzy control rules and learns how to control each process presented to it in real time, without the need for detailed process modelling. A suitable medical application to investigate this control strategy is atracurium-induced neuromuscular block (NMB) of patients in the operating theatre. Here, the patient response exhibits high non-linearity, and individual patient dose requirements can vary five-fold during an operating procedure. A portable control system was developed to assess the clinical performance of a simplified self-learning fuzzy controller in this application. A Paragraph (Vital Signs) NMB device monitored T1, the height of the first twitch in a train-of-four nerve stimulation mode. Using a T1 setpoint = 10% of baseline in ten patients undergoing general surgery, a mean T1 error of 0.45% (SD = 0.44%) is found while a 0.13-0.70 mg k-1 h-1 range in the mean atracurium infusion rate is accommodated. The result compares favourably with more complex and computationally-intensive model-based control strategies for the infusion of atracurium.