The application of GA-BP algorithm to intelligent diagnosis of coronary heart disease

A hybrid computational intelligent algorithm is proposed by combining the Back Propagation (BP) algorithm and the Genetic algorithm (GA), which avoids the defects that the results of traditional BP network are easily trapped into local optimum solutions. The network convergence rate is raised by using the iteration of Levenberg-Marquardt algorithm on the basis of the proposed GA-BP strategy. With the application of this method, the diagnosis rate can reach 98.98% after the simulation model of intelligent diagnosis is built based on the coronary heart disease cases of Cleveland Clinic Foundation and Hungarian Institute of Cardiology.