Cuckoo Search-Based Bayesian Networks for Medical Estimation System

This paper presents an estimation system. It depends on medical datasets for patient. It uses to medical diagnosis. The datasets are heart diseases and the nervous system diseases. Medical Estimation System is hybrid approaches between Bayesian Networks (BNs) and Cuckoo Search (CS). Modified- Full Bayesian Classifier (M-FBC) is one of algorithms in BNs. CS is a one of optimization algorithms. It uses easy technique and implementation. It has fewer control parameters and it could be easier than other optimization algorithms. So that hybrid with other algorithms to obtain ESCS the optimal structure. The Enhance Structure by Cuckoo Search (ESCS) obtains to optimization structure. In the input stage, the symptoms and the medical history for the patient are processed the structures. The proposal system to input all medical dataset and it filters and extracts the dataset. After of the structures, the proposal system can select the best optimal structure by trying to change the accepted accuracy by generation for each level in CS. The accuracy for M-FBC model is approximately (95.4%) for all diseases in the proposal system. While in The MFBC-ABC model, the accuracy is approximately (99.5%) for all diseases. The experimental results shown that the results for MFBC-ABC is better than on M-FBC. The accuracy for ESCS model is approximately (99.8%) for all diseases in the system, thus ESCS is the best model from all compared in this paper.