ScienceDirect Available online at

4. Conclusion 
This study uses GA based trained RFNN approach on UCI Cleveland dataset for diagnose the heart disease. The 
result of the approach was evaluated with RMSEs, sensitivities, specificities, precisions, F-scores, PMEs and 
accuracies of the training set, testing set and overall. The results of this study achieved a testing set (45 instances) 
accuracy of 97.78% and an overall accuracy of 96.63%. Compared with the ANN-Fuzzy_AHP approach where the 
accuracy is 91.1% for the testing set (45 instances), the proposed approach is distinctively better.
A future work may be considered to be jointly handled with medical experts to include different attributes that 
can affect the method’s decision making capabilities. Using different data sets from other sources may also be useful 
to test the proposed system performance. 
6
 
Kaan Uyar et al./ Procedia Computer Science 00 (2018) 000–000
References 
AHA, 2017. American Heart Association website. http://www.heart.org (11.05.2017) 
Aliev, R.A., Aliev R.R., Guirimov, B.G. ,Uyar K., 2007. Recurrent Fuzzy Neural Network Based System for Battery Charging. Lecture Notes in 
Computer Science 4492-II. 307-316. 
Aliev R.A., Aliev R.R., Gurimov B.G., Uyar K., 2008. Dynamic data mining technique for rules extraction in a process of battery charging, 
Applied Soft Computing 8 (3).1252-1258. 
Alsalamah, M., Amin, S., Halloran, J., 2014.Diagnosis of heart disease by using a radial basis function network classification technique on 
patients' medical records.2014 IEEE MTT-S International Microwave Workshop Series on RF and Wireless Technologies for Biomedical and 
Healthcare Applications (IMWS-Bio2014), London.1-4. 
Amiri A.M., Armano, G., 2013. Early Diagnosis of Heart Disease Using Classification And Regression Trees. The 2013 International Joint 
Conference on Neural Networks (IJCNN), Dallas, TX.1-4. 
Anooj, P.K., 2013. Implementing Decision Tree Fuzzy Rules in Clinical Decision Support System after Comparing with Fuzzy Based and Neural 
Network Based Systems. 2013 International Conference on IT Convergence and Security (ICITCS). 1-6. 
Bouktif, S., Hanna, E.M., Zaki, N., Abu Khousa, E., 2014. Ant Colony Optimization Algorithm for Interpretable Bayesian Classifiers 
Combination: Application to Medical Predictions. PLOS ONE 9(2).e86456, 12 pages. 
Çiftçi, F.B., Incekara, H., 2011. Web based medical decision support system application of Coronary Heart Disease diagnosis with Boolean 
functions minimization method. Expert Systems with Applications 38.14037–14043. 
Das, R., Turkoglu, I., Sengur, A., 2009. Effective diagnosis of heart disease through neural networks ensembles.Expert Systems with 
Applications 36. 7675–7680. 
De Falco, I., 2013. Differential Evolution for automatic rule extraction from medical databases.Applied Soft Computing 13.1265–1283. 
Hedeshi, N.G., Abadeh, M.S., 2014. Coronary Artery Disease Detection Using a Fuzzy-Boosting PSO Approach.Computational Intelligence and 
Neuroscience.Article ID 783734, 12 pages. 
Jabbar, M.A., Deekshatulu, B.L, Chndra, P., 2014. Alternating decision trees for early diagnosis of heart disease.Proc.of Int. Conf. on Circuits, 
Communication, Control and Computing (I4C 2014), MSRIT, Bangalore, India.322-328. 
Jabbar, M.A., Deekshatulu, B.L, Chandra P., 2015. Computational Intelligence Technique for early Diagnosis of Heart Disease. 2015 IEEE 
International Conference on Engineering and Technology (ICETECH), 20th March 2015, Coimbatore, TN, India.1-6. 
Kim, J.K., Lee, J.S., Park, D.K., Lim, Y.S., Lee, Y.H., 2014. Adaptive mining prediction model for content recommendation to coronary heart 
disease patients. Cluster Comput 17(3). 881-891. 
Miao K.H., Miao, J.H., Miao, G.J, 2016. Diagnosing Coronary Heart Disease Using Ensemble Machine Learning. (IJACSA) International 
Journal of Advanced Computer Science and Applications 7(10).30-39. 
Nahar, J., Imam, T., Tickle K.S., Chen Y.-P. P., 2013a. Computational intelligence for heart disease diagnosis: A medical knowledge driven 
approach. Expert Systems with Applications 40.96–104. 
Nahar, J., Imam, T., Tickle K.S., Chen Y.-P. P., 2013b. Association rule mining to detect factors which contribute to heart disease in males and 
females. Expert Systems with Applications 40.1086–1093. 
Nguyen, T., Khosravi, A., Creighton, D., Nahavandi, S., 2015.Medical data classification using interval type-2 fuzzy logic system and 
wavelets.Applied Soft Computing 30.812–822. 
NHLBI, 2017.National Heart, Lung, and Blood Institute website. https://www.nhlbi.nih.gov/health/health-topics/topics/hf (11.05.2017) 
Olaniyi, E.O., Oyedotun, O.K., Helwan, A., Adnan, K., 2015. Neural network diagnosis of heart disease. 2015 International Conference on 
Advances in Biomedical Engineering (ICABME), Beirut. 21-24. 
Paul, A.K., Shill, P.C., Rabin, M.R.I., Kundu, A., Akhand, M.A.H., 2015. Fuzzy membership function generation using DMS-PSO for the 
diagnosis of heart disease. 2015 18th International Conference on Computer and Information Technology (ICCIT), Dhaka. 456-461. 
Paul, A.K., Shill, P.C., Rabin M.R.I., Akhand, M.A.H., 2016. Genetic algorithm based fuzzy decision support system for the diagnosis of heart 
disease. 2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), Dhaka. 145-150. 
Polat, K., Güne
ş
, S., Tosun, S., 2006. Diagnosis of heart disease using artificial immune recognition system and fuzzy weighted pre-
processing.Pattern Recognition 39. 2186 – 2193. 
Reddy, G.T., Khare, N., 2017. An Efficient System for Heart Disease Prediction using Hybrid OFBAT with Rule-Based Fuzzy Logic Model. 
Journal of Circuits, Systems, and Computers 26 (4). 1750061-1-21 
Sagir, A.M., Sathasivam, S., 2017. A Novel Adaptive Neuro Fuzzy Inference System Based Classification Model for Heart Disease Prediction. 
Pertanika J. Sci. & Technol. 25 (1). 43 – 56. 
Samuel, O.W., Asogbon , G.M., Sangaiah, A.K., Fang, P. , Li, G., 2017. An integrated decision support system based on ANN and Fuzzy_AHP 
for heart failure risk prediction. Expert Systems With Applications 68. 163–172. 
Shaoa, Y.E., Houa, C.-D., Chiuba C.-.C., 2014. Hybrid intelligent modeling schemes for heart disease classification. Applied Soft Computing 
14.47–52. 
UCI, 1990.Heart disease dataset. http://archive.ics.uci.edu/ml/machine-learning-databases/heart-disease/ (10.02.2017) 
Uyar, K., 2006. Modeling and Simulation of NiCd Batteries Behaviour Under Fast Charging with Genetic Algorithm Based Trained Recurrent 
Fuzzy Neural Network. Proc. of the 7th Int. Conf. on Application of Fuzzy Systems and Soft Computing (ICAFS2006), Siegen, Germany. 
247-255 
WHO, 2017. World Health Organization, Media centre, cardiovascular diseases fact sheet webpage. 
http://www.who.int/mediacentre/factsheets/fs317/en/ (12.05.2017)

---

 
Kaan Uyar et al. / Procedia Computer Science 120 (2017) 588–593 
593
 
Kaan Uyar et al./ Procedia Computer Science 00 (2018) 000–000 
5
Precision
Recall
Precision
*
ecall
R
Score
F



*
2
(5) 
FN
TN
FP
TP
FN
FP
PME





(6) 
FN
TN
FP
TP
TN
TP
Accuracy





(7)