provide a diagnostic model using a combination of two neural network algorithms and a genetic algorithm

introduction: improvement of technology can increase the use of machine learning algorithms in predicting diseases. early diagnosis of the disease can reduce mortality and morbidity at the community level. material and methods: in this paper, a clinical decision support system for the diagnosis of gestational diabetes is presented by combining artificial neural network and metaheuristic algorithm. in this study, four metainnovative algorithms of genetics, ant colony, particle swarm optimization and cuckoo search were selected to be combined with artificial neural network. then these four algorithms were compared with each other. the data set contains 768 records and 8 dependent variables. this data set has 200 missing records, so the number of study records was reduced to 568 records. results: the data were divided into two sets of training and testing by 10fold method. then, all four algorithms of neuralgenetic network, antneural colony network, neural networkparticle swarm optimization and neural networkcuckoo search on the data the trainings were performed and then evaluated by the test set. and the accuracy of 95.02 was obtained. also, the final output of the algorithm was examined with two similar tasks and it was shown that the proposed model worked better. conclusion: in this study showed that the combination of two neural network and genetic algorithms can provide a suitable predictive model for disease diagnosis.