H∞ control analysis of patient agitation management in the critically ill

  • Authors:

  • J. Geoffrey Chase;Andrew D. Rudge;Dominic S. Lee;Geoffrey M. Shaw

  • Affiliations:

  • Department of Mechanical Engineering, Centre for Bioengineering, University of Canterbury, Private Bag 4800, Christchurch, New Zealand.;Department of Mechanical Engineering, Centre for Bioengineering, University of Canterbury, Private Bag 4800, Christchurch, New Zealand.;Department of Mathematics and Statistics, Centre for Bioengineering, University of Canterbury, Private Bag 4800, Christchurch, New Zealand.;Department of Intensive Care Medicine, Christchurch Hospital, Christchurch, New Zealand

  • Venue:
  • International Journal of Intelligent Systems Technologies and Applications
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Patient agitation in the critically ill damages patient health,increases length of stay, and contributes to rising healthcarecosts. Recently developed quantifiable measures of patientagitation create the potential for automated sedation management. Aphysiologically verified non-linear model of agitation-sedationdynamics is used to design and evaluate sedative infusioncontrollers. A simplified linear model is used for H∞analysis and control design to minimise the transfer functionbetween unknown stimulus and resulting patient agitation. Regionsof stability and instability are identified usingHï¾∞ Lyapunov methods, and bolus-based heavyderivative feedback control is shown to best minimise theHï¾∞ norm. Hï¾∞ controllersyield an approximately 25% improvement over prior choices, and anearly 50% improvement over current clinical practice. Importantly,these results show that general trends in the linear system arestill reflected.