NestEn_SmVn: boosted nested ensemble multiplexing to diagnose coronary artery disease

Coronary artery disease (CAD) is the most prominent disease that is responsible for increasing mortality and morbidity rate from past few decades. Early and accurate detection of CAD (a type of cardiovascular diseases) is among the most pressing needs of society. In this research work, experiments h...

Full description

Saved in:
Bibliographic Details
Published in:Evolving systems 2022-04, Vol.13 (2), p.281-295
Main Authors: Shastri, Sourabh, Singh, Kuljeet, Kumar, Sachin, Kour, Paramjit, Mansotra, Vibhakar
Format: Article
Language:English
Subjects:
Artificial Intelligence
Complex Systems
Complexity
Engineering
Original Paper
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Coronary artery disease (CAD) is the most prominent disease that is responsible for increasing mortality and morbidity rate from past few decades. Early and accurate detection of CAD (a type of cardiovascular diseases) is among the most pressing needs of society. In this research work, experiments have been carried out with Cleveland dataset in four phases such as ( i ) single classifiers, ( ii ) boosted stacking nested ensemble, ( iii ) boosted voting nested ensemble, and ( iv ) boosted stacked voting nested ensemble. A generalized framework NestEn _S m V n has been proposed for designed nested ensemble models (phases ii to iv above). The proposed framework ( NestEn _S m V n ) using boosted stacked voting nested ensemble learning techniques having model (ID E ID3 -G ID6 ) designed with adaptive boosting and Bayesian network as base-classifiers along with SMO and LMT as meta learners that achieved an highest accuracy of 98.68% with F-measure and ROC values of 98.70 and 99.00% respectively. The best proposed model (ID E ID3 -G ID6 ) from nested ensemble (phase iv ) using proposed framework ( NestEn _S m V n ) has outperformed all other models from phases ( i-iv ) and all previous works. Our proposed framework can support the clinical decision system and is able to replace previous CAD diagnostic techniques.
ISSN:1868-6478
1868-6486
DOI:10.1007/s12530-021-09384-3