phat 18: e8

Research Article

Prediction of Heart Disease using Biomedical Data through Machine Learning Techniques

Download40 downloads
  • @ARTICLE{10.4108/eai.30-8-2021.170881,
        author={Nagaraj M. Lutimath and Neha Sharma and Byregowda B K},
        title={Prediction of Heart Disease using Biomedical Data  through Machine Learning Techniques},
        journal={EAI Endorsed Transactions on Pervasive Health and Technology: Online First},
        volume={},
        number={},
        publisher={EAI},
        journal_a={PHAT},
        year={2021},
        month={8},
        keywords={Machine Learning, Random Forest, Python, Bio-Medical data, Information Retrieval, Predictive Analysis},
        doi={10.4108/eai.30-8-2021.170881}
    }
    
  • Nagaraj M. Lutimath
    Neha Sharma
    Byregowda B K
    Year: 2021
    Prediction of Heart Disease using Biomedical Data through Machine Learning Techniques
    PHAT
    EAI
    DOI: 10.4108/eai.30-8-2021.170881
Nagaraj M. Lutimath1, Neha Sharma2,*, Byregowda B K3
  • 1: Associate Professor, Department of Information Science and Engineering, East Point College of Engineering and Technology, Bangalore, India
  • 2: Assistant Professor, Department of AIT Computer Science and Engineering, Chandigarh University, Punjab, India
  • 3: Assistant Professor, Department of Information Science and Engineering, Sir M Visveswaraya Institute of Technology, Bengaluru, India
*Contact email: nehasharma0110@gmail.com

Abstract

INTRODUCTION: Random Forests are an important model in machine learning. They are simple and very effective classification approach. The random forest identifies the most important features of a given problem.

OBJECTIVES: The heart disease is cardiovascular disease, with a set of conditions affecting the heart. During heart disease, there will be heartbeat problems with congenital heart disorders and coronary artery defects. A coronary heart defect is a heart disease, which decreases the flow of blood to the heart. When the flow of blood decreases heart attack occurs. It is necessary to analyse the prediction of heart attack based on the symptoms.

METHODS: The available data set of patients with heart defects symptoms is taken and analysed in this paper using the random forest and decision tree regression models. The missing data is updated using mean value of the attribute. Python language is used to predict the accuracy.

RESULTS: Three performance measures are taken for analysing the available UCI Cleveland data set for heart disease. The performance measures are the Mean Absolute Error, Mean Squared Error and Root Mean Squared Error. Vital attributes of the data set are taken for analyses using the random forest regression model and decision tree regression model. The analyses shows that the slope attribute provides the better prediction for the heart disease. The results are shows that the females are more prone to heart attack.

CONCLUSION: Prediction of heart disease using the UCI machine learning data set at Cleveland repository is analysed using random forest regression model and decision tree regression models. We find random forest regression model provides better accuracy than decision tree regression model.