International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 09 Issue: 07 | July 2022
www.irjet.net
p-ISSN: 2395-0072
E-Healthcare monitoring System for diagnosis of Heart Disease using Machine Learning Rakshitha G R1, Shejal Shankar2, Vijay Kumar S3 1Student, Dept. of Information Science and Engineering, BNM Institute of Technology, Karnataka, India
2 Student, Dept. of Information Science and Engineering, BNM Institute of Technology, Karnataka, India 3 Assistant Professor, Dept. of Information Science and Engineering, BNM Institute of Technology,
Karnataka, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract
– Modern healthcare technology advancements have a tremendous impact on how well medical services are provided and how many lives are saved. Cardiovascular illness, often known as heart disease, is the most deadly and complex disease and is difficult to diagnose with the unaided eye. According to the WHO cardiovascular disease (CVD) claims millions lannually. According to Global Burden of Disease, CVDs account for about 24.8% of all deaths in India. The number of CVD-related deaths in India has increased annually, from 2.26million in 2019 to 4.77 million in 2020. Heart disease is typically diagnosed by a doctor after reviewing the patient's medical history, the results of their physical exam, and any troubling symptoms. However, the results of this method of diagnosis do not reliably identify patients who have cardiac disease. Additionally, it is costly and computationally challenging to assess. Most clinical diagnoses are made by doctors with training and experience. However, incidences of incorrect diagnosis and treatment continue to be reported. Numerous diagnostic tests are required of patients. Many times, not every test helps with a disease's accurate diagnosis. Machine learning algorithms can foresee this kind of illness. Utilizing various machine learning approaches to quickly analyze and diagnose HD is one of the project's primary research goals. Additionally, it appears that the machine learning prediction model is a crucial feature in this field of study. With the use of certain methods like Feature Selection, Record, Attribute Minimization, and Classification, this work intends to offer a new heart disease prediction model in this situation.
over time by exploiting data and experience. Along with being capable of supervised learning, unsupervised learning, and reinforcement learning, ML and DL models can also learn in other ways. To categorize or generate predictions, supervised learning uses labelled datasets; this process necessitates human interaction to accurately identify the incoming data. Unsupervised learning, in contrast, does not require labelled datasets; instead, it finds patterns in the data and groups them according to any differentiating traits. A model learns to improve its accuracy for completing an activity in an environment based on feedback through the process of reinforcement learning. Machine learning is becoming increasingly popular as a result of its superior accuracy when taught on massive amounts of data. When it comes to challenging issues like speech recognition, natural language processing, and image classification, it really excels.
Key Words: Feature Selection, Classification, Machine Learning.
In [4], offers to use data mining techniques to uncover hidden patterns that are significant to cardiac illnesses and to forecast whether the patient has a bad heart. The Weka tool is used to implement the selected strategy. It has tools for pre- processing information, categorization, regression, clustering, association rules, and visualization. For this study, decision tree classifiers with 10-fold Cross-Validation in Test mode aretaken into consideration.
2. RELATED WORK In [3], introduces a method that uses fewer criteria to predict the presence of heart disease more effectively. Thirteen factors were first taken into consideration for predicting heart disease. In our research, a genetic algorithm is utilized to identify the characteristics that are most helpful in the detection of cardiac conditions. Using genetic search, thirteen traits are reduced to six. The diagnosis of patients is then predicted using three classifiers—Naive Bayes, Classification by Clustering, and Decision Tree—with the same accuracy as before the reduction in the number of characteristics. The observations also show that after including feature subset selection the Decision Tree data mining technique beats the other two data mining strategies.
1. INTRODUCTION Systems can learn and develop automatically thanks to machine learning, a subset of AI. Its the study of computer algorithms that can get better on their own
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 1592