A Hybrid Heart Disease Prediction System Using Evolutionary Datamining Algorithms

Today's Modern hospitals contain physicians, patients and clinical workers as well as different procedures, including the patient's treatment. As of late current frameworks and procedures have been acquainted in health-care organizations with encourage their tasks. A large amount of clinical records are put away in databases and data warehouses. Such databases and applications vary from each other. The essential ones store just essential data about patients, for example, name, age, address, blood classification, and so on. The further developed ones let the hospital staff register patients' visits and store point by point data about their health condition. A few systems likewise encourage patients' enrolment, units' funds and planning of visits. Recently a new type of a medical system has emerged: medical decision support system. It originates in the business intelligence and is to support medical decisions. The data which is stored in such a system may contain valuable knowledge hidden in medical records. Appropriate processing of this information has potential of enriching every medical unit by providing it with experience of many specialists who contributed their knowledge to building the system. The situation described above is the reason for a close collaboration between computer scientists and medical staff. It aims at development of the most suitable method of data processing which would enable discovering nontrivial rules and dependencies in data. The results may improve the process of diagnosing and treatment as well as reduce the risk of a medical mistake or the time of a diagnosis delivery. This may turn out to be critical especially in emergency incidents. The research area which seeks for methods of knowledge extraction from data is called knowledge discovery or data mining. It utilizes various data mining algorithms to analyze databases. The main objective of this study was to develop a Hybrid Learning Algorithm combining Cascaded Neural Network and Genetic Algorithm (GBCNN) for predicting the threat of heart disease to a patient with the medical records got from the patients. GBCNN had the different attributes that it did not use already determined number of hidden units, but the hidden units got summed with one another till the error was decreased using Genetic Algorithm. By exploiting this distinct feature of the GBCNN, an automated algorithm for prediction was built which would accurately predict heart disease and was also efficient. With the appropriate version of GBCNN classifiers, this technique could hence develop a best possible amount of hidden units for a given architecture. As one part of the study the researcher had studied five algorithms: C4.5, Multilayer Perceptron Neural Network, Na鴳e Bayes, Cascaded Neural Network and Genetic Algorithm. For the evaluation UCI database was used. Several performance metrics were utilized: percent of correct classifications, True/False Positive rates, AUC, Precision, Recall, F-measure and a set of errors. The evaluation of effectiveness and accuracy of data mining methods for these algorithms showed that the most accurate method was Cascaded Neural Network, next was Na鴳e Bayes Classifier, next the Multilayer Perceptron Neural Network and finally C4.5 algorithm. Then the scholar had used the Genetic Algorithm to be combined with Cascaded Neural Network to build a hybrid heart disease prediction system. The experimental results have proved that the proposed approach has achieved improvement in accuracy. This implements that the Cascaded Neural Network combined with genetic algorithm will be a sought-after classifier that could be useful guide for the doctors to efficiently and accurately predict the heart disease with less time.