Adaptive predictive control using a neural network in acute hypotension

Abstract

We examined a mean arterial blood pressure (MAP) control during acute hypotension using a neural network (NN). A proportional-integral-derivative (PID) control, an adaptive predictive control using a NN (APCNN), and a combined control of APCNN and PID (APCNN-PID) were tested. In anesthetized rabbits, using the NE of 25 µg ml-1, the PID control, APCNN and APCNN-PID prevented severe hypotension. Using NE of 50 µg ml-1, the rabbits recovered from acute hypotension for all systems tested but showed sustained MAP oscillation during PID control. Utilization of a NN for adaptive predictive control systems could facilitate the development of a drug infusion apparatus.