Application And Research Of Fuzzy Genetic Algorithm In Medication Aid In Decision Making

In recent years, with the development of medical and public health community, drug use in clinical activities will gradually become the public focus of attention. Rational drug use means that it is to select best drugs and preparations according to disease types, patient conditions and pharmacology theory, to formulate or adjust programme aim at effective, safe, economic measures for prevention and cure of disease. The guiding principle to solve technical problems is evidence-based medicine. The main related research methods contain randomized controlled study, system assessment, clinical guidelines, and so on.Genetic algorithm (GA) is random search algorithm that simulatesbiological evolution mechanisms such as natural selection and "survival of the fittest". It's oneof the most effective methods to solve complex problems.Theory of genetic algorithm is studied, and the affect of genetic parameters to the evolutionary process is analyzed. A fuzzy genetic algorithm is proposed to improve performances of genetic algorithm. The crossover rate and mutation rate are adjusted according to the evolutionary situation on-line. The differences between the maximum fitness with the average fitness and change of the fitness in successive generations are chosen as the inputs, and fuzzy inference rules are set up by OMGA and experience. Example shows that the fuzzy genetic algorithm accelerates the evolutionary process and avoids premature convergence effectively. The fuzzy genetic algorithm also improves the optimum value.Drug grouping problem is an important part of prescriptions specific to diseases, and the fuzzy genetic algorithm proposed by the paper has been used in drug grouping problem. According to the characteristics of drug grouping problem, the algorithm put up direct coding schemes, set empty encoding in the last bit, studied proper population initialization methods to determine the cross and variation rules fit for direct coding .So the constraint conditions of grouping disappeared in such rules to prevent the loss of a good prescription in a large extent.