Robust Control of Maintenance-Phase Anesthesia

In biomedical systems, feedback control can be applied whenever adequate sensors, actuators, and sufficiently accurate mathematical models are available. The key issue is the capacity of the control algorithm to tackle the large levels of uncertainty, both structured and unstructured, associated with patient dynamics. In the particular case of intravenous anesthesia considered here, manipulated variables are drug infusion rates, administered by syringe pumps, and the measured signal outputs are the levels of hypnosis or depth of anesthesia (DoA) and of neuromuscular blockade (NMB). Figure 1 provides an example of a loop closed for the control of NMB.