International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 05 | May 2022
p-ISSN: 2395-0072
www.irjet.net
PREDICTION OF HEART DISEASE USING LOGISTIC REGRESSION Kavya Sreehari¹, Devika Santhosh Kumar², Muhammed Shameem S³, Sumeesh S⁴, Bismi M⁵ ¹,²,³,´ UG Scholar, Department of Computer Science and Engineering, µAsst. Prof, Department of Computer Science and Engineering, UKF College of Engineering and Technology, Kerala, India -----------------------------------------------------------------------------***------------------------------------------------------------------------Abstract: Heart disease symptoms are caused by abnormal heartbeats and diseased heart muscle. There are two cases of
prediction of which the correct prediction can help to prevent threats whereas incorrect prediction can lead to fatal. Here, the paper is based on heart disease prediction using logistic regression model. Machine Learning is one the trending technology in which various researches around the world is used for predicting diseases. Nowadays it’s important for the early detection and for its treatment.The dataset consists of 14 attributes used for performing the analysis. Accuracy is validated and promising results are achieved. Heart disease dataset analysed is used to predict the result whether the patient has heart disease or not i.e., using logistic regression technique. This prediction gives the result in the form of logistic representations which produces efficient and accurate results in healthcare sectors. Keywords: Heart disease, Machine learning, Logistic regression
1. INTRODUCTION According to WHO, 17.9 million people year die due to heart related diseases. There are different types of heart disease, some are preventable. Heart disease refers to any condition affecting the heart and blood vessels. There are different types of heart disease affecting the heart and blood vessels in different ways. Early detection of cardiovascular disease may help to avoid complications. Unhealthy lifestyle habits are reason for increase in heart related diseases diet high in saturated fats, trans fats, and cholesterol has been linked to heart disease and related conditions, such as atherosclerosis. Excess salt in the diet can increase blood pressure. Some risk factors for heart disease can be controlled whereas some cannot be controlled, for example family history. One of the main risk factor for heart disease is high blood pressure. It is often called a “silent killer” as it usually shows no symptoms. Early detection of heart diseases is required to reduce the health complications. In healthcare sector machine learning has been used in diagnosing and predicting various diseases using different models. The study intends to find the most important predicators of heart diseases and predicting the overall risk by using logistic regression. Healthcare expenses are overwhelming national and corporate budgets due to asymptomatic diseases including heart diseases. Therefore, there is an immediate need to detect and treat such diseases. Coronary heart disease is also known as ischemic heart disease. Plaque build-up in the walls of the arteries leads to coronary heart disease. Some of the symptoms of heart disease include pain in the chest, breath shortness, and discomfort in the arms, back, neck, jaw, or stomach.
2. RELATED WORKS Machine learning is widely used in almost all fields including healthcare sector. Most machine learning algorithms alive concerned with discovering interrelationship between datasets. Once Machine Learning Algorithms can identify on certain correlations, the model can either use these relationships to predict future observations or generalize the info to reveal interesting patterns. In Machine Learning there are various types of algorithms like Regression, Linear Regression, , Naive Bayes Classifier, Bayes theorem, Logistic Regression KNN (K-Nearest Neighbour Classifier), Decision Tress, Entropy, ID3, SVM (Support Vector Machines), K-means Algorithm, Random Forest .Lots of research work have been done for assessment of the classification accuracies of different machine learning algorithms by using the Cleveland heart disease database which is uninhibitedly accessible at an online data mining repository of the UCI. Authors Bayu Adhi Tama, Afriyan Firdaus, Rodiyatul
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 1837