| چکیده انگلیسی مقاله |
Aim Data mining focuses on the process of discovering interesting, useful patterns and relationships in large volumes of data in order to generate new knowledge from large databases. It has been applied in the medical domain such as heart disease affluently. Heart disease still remains the main cause of death worldwide. According to the statistic data from WHO, one third population worldwide died from heart disease. Here, we have reviewed the application of data mining methods for prediction of heart disease. Method This is a narrative review. We searched the IEEE, Science Direct, Google Scholar, Web of Science, Scopus, Elsevier, MEDLINE database through PUBMED in Aug 2017 and out of 95 resulted papers for all results up to April 2017 obtained; 34 ones had the inclusion criteria and their full texts were accessible. Result Results of the studies have been considered to evaluate based on two table: The first table consists of 13 items: Title, Authors Name, year of the published, type of study, the name of paper, risk factors, types of data mining methods, types of Algorithms, sample size, Output quality, The type of system used, Type of dependent variable. The second table consists of 2 items: Features of study (dependent variable and independent variable). In a division other 21 studies have been done on Heart disease, 3 studies on Coronary Heart disease. ,5 studies on Cardio Vascular Disease, 1 studies on Heart Attack, 4 studies on Coronary Artery Disease, 2 studies on cardio Metabolic Risk , 1 studies on Heart and Diabetes Diseases, 1 studies on heart disease, thyroid, diabetes, and hepatitis. Conclusion The main focus of data mining methods application for heart disease has been during the years of 2008 to 2017. Our analysis Of the 34 articles showed the best algorithms based on output accuracy are Neural Networks, Decision Trees, and Naive Bayes with accuracy of 100%, 99.62%, and 90.74% respectively. Our analysis shows that among these three classification models backpropagation neural network Neural Networks have predicted Heart disease with highest accuracy. Aim Data mining focuses on the process of discovering interesting, useful patterns and relationships in large volumes of data in order to generate new knowledge from large databases. It has been applied in the medical domain such as heart disease affluently. Heart disease still remains the main cause of death worldwide. According to the statistic data from WHO, one third population worldwide died from heart disease. Here, we have reviewed the application of data mining methods for prediction of heart disease. Method This is a narrative review. We searched the IEEE, Science Direct, Google Scholar, Web of Science, Scopus, Elsevier, MEDLINE database through PUBMED in Aug 2017 and out of 95 resulted papers for all results up to April 2017 obtained; 34 ones had the inclusion criteria and their full texts were accessible. Result Results of the studies have been considered to evaluate based on two table: The first table consists of 13 items: Title, Authors Name, year of the published, type of study, the name of paper, risk factors, types of data mining methods, types of Algorithms, sample size, Output quality, The type of system used, Type of dependent variable. The second table consists of 2 items: Features of study (dependent variable and independent variable). In a division other 21 studies have been done on Heart disease, 3 studies on Coronary Heart disease. ,5 studies on Cardio Vascular Disease, 1 studies on Heart Attack, 4 studies on Coronary Artery Disease, 2 studies on cardio Metabolic Risk , 1 studies on Heart and Diabetes Diseases, 1 studies on heart disease, thyroid, diabetes, and hepatitis. Conclusion The main focus of data mining methods application for heart disease has been during the years of 2008 to 2017. Our analysis Of the 34 articles showed the best algorithms based on output accuracy are Neural Networks, Decision Trees, and Naive Bayes with accuracy of 100%, 99.62%, and 90.74% respectively. Our analysis shows that among these three classification models backpropagation neural network Neural Networks have predicted Heart disease with highest accuracy. |
