Based On Bayesian Probability Of Flu Vaccine

Current, the market of influenza vaccine is a triple vaccine, but this type of fluvaccine manufacturing cost is high. In order to solve this problem, we consider whetheryou can use the two assembly of influenza vaccine. Here we use Bayesian probability, aBayesian forecasting model, to predict the probability of four species in winter2009fluvirus subtypes in various countries within the scope of, select two kinds of maximumprobability virus subtype probability self are calculated, and then the virus subtype tomanufacture two mounted influenza vaccine; in addition, consider to two mounted vaccinetreatment effect may be lower than the triple vaccine, we established a comprehensiveevaluation model; finally, in order to half that virus forecast2009the biggest threat, so thatexperts can determine the vaccine components, we establish the virus screening model here,screening out the biggest threat virus type with this model.When using loading bigeminal vaccine, in order not to reduce the curative effect ofloading bigeminal vaccine, need in each country in the winter of2009to predict theprobability of the presence of the virus subtype, using bayes prediction model isestablished, we may safely draw the various countries in the winter that two kind of virussubtype in most probability, and thus determine the vaccine on plan, specific delivery planto see in table7. Bigeminal vaccine relative triple vaccine, though rising vaccine coverage,but vaccination success rate has fallen, in order to consider the two factors, we establishedthe vaccine effect evaluation model, in order to draw a second league outfit and the joint ofthe vaccine comparison results. Finally, in order to facilitate virus experts identified andproduction, we established a virus screening model, using the antigen in the modelsimilarity matrix and antigen utility, select the most effective antigen with matlabprogramming.