• DocumentCode
    391061
  • Title

    Sufficient conditions for optimality of controls in biomedical systems

  • Author

    Ledzewicz, Urszula ; Schättler, Heinz

  • Author_Institution
    Dept. of Math. & Stat., Southern Illinois Univ., Edwardsville, IL, USA
  • Volume
    3
  • fYear
    2002
  • fDate
    10-13 Dec. 2002
  • Firstpage
    3524
  • Abstract
    A general class of optimal control problems of Bolza type is considered which arise in mathematical models of biomedical systems when chemotherapy treatment protocols for a disease over a fixed interval are considered. The controls represent the dosages of drugs administered and in this paper an objective which is quadratic in the controls is analyzed. This choice leads to continuous optimal controls which alternate between values in the interior and on the boundary of the control set. Using the method of characteristics a local field of extremals is constructed around a reference trajectory and sufficient conditions for strong local minima are given. These conditions apply to several biomedical models of chemotherapy for diseases like cancer or HIV infections which have been considered in the literature before, but not in the context of sufficient conditions. The result is illustrated with a simulation of a three-compartment model for cancer chemotherapy.
  • Keywords
    biomedical engineering; optimal control; protocols; Bolza type; HIV infections; biomedical systems; chemotherapy treatment protocols; mathematical models; optimal control problems; optimality of controls; reference trajectory; strong local minima; sufficient conditions; Cancer; Context modeling; Control systems; Diseases; Drugs; Mathematical model; Medical treatment; Optimal control; Protocols; Sufficient conditions;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Decision and Control, 2002, Proceedings of the 41st IEEE Conference on
  • ISSN
    0191-2216
  • Print_ISBN
    0-7803-7516-5
  • Type

    conf

  • DOI
    10.1109/CDC.2002.1184421
  • Filename
    1184421