Abstract: In view of the current medical data mining process, the medical data mining problems, proposes a mining algorithm based on association rules of medical data, searching through frequent itemsets of medical data, generate strong association rules, the calculated average value and standard deviation of large medical data sets, implementation the association rules adjustment, determine the medical data attributes, construct the appropriate fitness function, according to the classification attribute combination value of fitness function selection, crossover and mutation reduction, calculation of information entropy on classification attribute reduction, classification and attribute information gain by calculating the expected information, the establishment of decision tree mining model, implementation of medical data mining. The experimental results show that the proposed algorithm can effectively improve the level of medical data preprocessing, and the effect of data mining is better.