Prediction of Thalassemia Based on SAELM Hybrid Algorithm

Thalassemia is a serious hereditary disease, which cannot be cured. It's valuable to study on thalassemia. The input weights and biases of Extreme Learning Machine (ELM) can be randomly initialized, which sometimes cannot get the better result. So a new algorithm named SAELM based on thalassemia is proposed by using the strong global optimization ability of Simulated Annealing (SA) algorithm, which can find the best weights and biases of ELM algorithm. Simulation results illustrated that the SAELM is better than ELM in the main evaluation indices, so SAELM algorithm can be used as a medical reference index in the screening of thalassemia.