Abstract
Manual medical diagnosis which depends on physicians’ knowledge to diagnose the presence of the symptoms of the disease is impracticable. Therefore, automatic and intelligent medical diagnosis has become very useful to the physicians when dealing with huge amount and high dimensional medical database. In this paper, we have proposed hybridization method by improving MLP learning with Biogeography Based Optimization (BBO) to be adopted and applied in five medical diagnoses. Comparisons are done between the following proposed methods: hybrid Particle Swarm Optimization (PSO) and MLP; hybrid Genetic Algorithm (GA) and MLP; and hybrid Artificial Fish Swarm Algorithm (AFSA) and MLP using the same standard parameters. Results are analyzed in terms of their classification accuracy. The performance of each method was evaluated based on their specificity, sensitivity, accuracy and precision. The findings disclose that BBO is a promising optimization tool in enhancing MLP learning with better average accuracy and convergence rate in intelligent medical diagnosis.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have