Page 44 - Control Theory in Biomedical Engineering
P. 44

Modeling and control in physiology  31


              glucose concentration. Thus, the adaptive control algorithm can settle this
              type of challenge (Turksoy and Cinar, 2014; Nath et al., 2018).
              •  Adaptive control problem in convergence eye movements
              The oculomotor human system contains two adaptive controllers in order to
              have clear and comfortable binocular vision. Thus, the oculomotor human
              system has a natural and autonomous adaptive capacity regulated by neural
              mechanisms (Erkelens et al., 2020).

              3.4.3 Fuzzy logic control
              In nature, most systems and concepts are naturally unpredictable and fuzzy,
              hence the importance of fuzzy set theory in real problems and especially in
              human physiology. In fact, fuzzy logic control is adequate for medicine
              because it is tolerant of peculiar imprecision. The fuzzy concept is based
              on fuzzy rules of the form IF … THEN. A fuzzy logic controller is equiv-
              alent linguistically to a PI controller (Mahfouf et al., 2001).
                 In the literature, fuzzy logic controllers are the most incorporated control
              algorithms in medicine applications. Some of them are described in the par-
              agraphs that follow.
              •  Fuzzy logic control problem in anesthesia
              Fuzzy logic controllers are widely adapted and used in anesthesia (Derrick
              et al., 1998). Generally, anesthetists fix the rules of this type of controller
              since they have extensive experience with patients. One example of these
              rules is IF “blood pressure is decreased” THEN “reduce drug infusion.”
              More sophisticated rules are also composed.
              •  Fuzzy logic control problem in blood glucose regulation
              Fuzzy logic controller is also used to regulate blood glucose in type 1 diabetic
              patients. This type of controller is based on zero order Takagi-Sugeno fuzzy
              logic architecture (Nath et al., 2018). This controller is characterized by two
              inputs and one output. The inputs are the error and the derivative in sub-
              cutaneous glucose concentration. The output of the fuzzy controller is the
              exogenous insulin infusion rate.



              4 Future trends and challenges
              One of the more complex problems is biofeedback therapy. It is considered
              as a contemporary challenging in preventive healthcare researches. In fact,
              biofeedback is a mind–body training technique that involves using visual
              or auditory feedback to gain control over involuntary bodily functions
              (Patcharatrakul et al., 2020). This may include gaining voluntary control
   39   40   41   42   43   44   45   46   47   48   49