Clinical decision support model for prevailing diseases to improve human life survivability

Constantly increasing amount of heterogeneous prevailing disease patient data can redefines medical research and clinical practice for human life survival. Computational intelligent techniques help to translate them into knowledge base that is applicable in health-care. Prediction of such diseases at early stages is biggest challenge for doctors in the country. Previous studies on prevailing diseases focus on individual diseases rather than many with similar symptom. Few of these models have constraints in finding good parameters with high accuracy. The proposed clinical decision support system in this paper models the patient´s diseases state statically from his heterogeneous data to reveal the correct diagnosis by formalizing the hypothesis based on test results and symptoms of the patient before recommending treatments for the prevailing diseases. Its goal is to assist clinician in diagnosing the patient by analyzing his available data and relevant information. The proposed model designed using data mining techniques such as neural network, decision tree, statistical method, Naive Bayes, classifier and clustering pattern analysis for improving human life survivability. Several clinical data-set are used to evaluate and demonstrate the proposed model for early prediction of prevailing disease.