Ensemble technique to predict heart disease using machine learning classifiers

The exact forecast of heart disease is necessary to proficiently treat cardiovascular patients before a heart failure happens. Assuming we talk about AI techniques can be accomplished utilizing an ideal AI model with rich medical services information on heart diseases. To begin with, the feature ext...

Full description

Saved in:
Bibliographic Details
Published in:Network biology 2023-03, Vol.13 (1), p.1-16
Main Author: Chaurasia, Aparna
Format: Article
Language:English
Subjects:
Cardiovascular disease
Cardiovascular diseases
Classifiers
comparison model
Congestive heart failure
Datasets
Feature extraction
feature selection
Health services
Heart
heart disease
Heart diseases
Illnesses
Machine learning
Mathematical models
sfs
stacking model
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The exact forecast of heart disease is necessary to proficiently treat cardiovascular patients before a heart failure happens. Assuming we talk about AI techniques can be accomplished utilizing an ideal AI model with rich medical services information on heart diseases. To begin with, the feature extraction technique, gradient boosting-based sequential feature selection (GBSFS) is applied to separate the significant number of features (5, 7, 9, and 11) from coronary illness dataset to create important medical services information. The stacking model is prepared for coronary illness forecast. A comparison model is made between datasets with prominent features (5, 7, 9, and 11) as well as all features. The proposed framework is assessed with coronary illness information and contrasted and customary classifiers in view of feature elimination include determination strategies. The proposed framework acquires test accuracy of 98.78%, which is most noteworthy in marking model with 11-featuers and higher than existing frameworks. This outcome shows that our framework is more powerful for the expectation of coronary illness, in contrast with other cutting edge strategies.
ISSN:2220-8879