An Articulate Heart Attack Detection System Using Mine Blast Optimization (MBO) Based Multilayer Perceptron Neural Network (MLPNN) Model

Rajesh Pandian N; Shanthi D; Selvaganesh N

doi:10.52866/ijcsm.2023.02.02.012

Authors

Rajesh Pandian N Department of Computer Science and Engineering, Faculty of Engineering and Technology, Jain (Deemed-to-be University), Bengaluru, Karnataka-560069. India.
Shanthi D Department of Computer Science and Engineering, PSNA College of Engineering and Technology, Kothandaraman Nagar, Dindigul, Tamil Nadu 624622, India https://orcid.org/0000-0001-8722-424X
Selvaganesh N Department of Information Technology, Sri Venkateswara College of Engineering, Sriperumbudur, Tamil Nadu 602117, India https://orcid.org/0000-0001-5685-2438

DOI:

https://doi.org/10.52866/ijcsm.2023.02.02.012

Keywords:

Heart Attack Detection, Machine Learning Model, Big Data, Regression based Preprocessing, Mine Blast Optimization (MBO), and Multi-Layer Perceptron Neural Network (MLPNN).

Abstract

The creation of an automated system for heart disease detection was once one of the more common
undertakings in the healthcare industries. For this purpose, the different types of big data analytics technologies are
developed in the conventional works to predict the heart disease. Still, it limits with the problems associated to the
elements of high complexity, time consumption, over fitting, and mis-prediction results. Because the previous
methods did not optimize the best features, they did not give accurate results in heart attack detection, so the
system is needed to control the death ratio.Therefore, the proposed work objects to implement a novel Mine Blast
Optimization (MBO) based Multi-Layer Perceptron Neural Network (MLPNN) technique to predict the heart
attack from the given datasets. The proposed detection framework includes the stages of preprocessing, feature
optimization, and classification. Here, the regression based preprocessing model is implemented to normalize the
attributes for increasing the quality. Then, the MBO technique is also used to choose the relevant features based on
the best optimal solution. It also helps to reduce the increase the training of classifier with reduced time
consumption and high detection accuracy. Finally, the MLPNN technique is utilized to predict the classified label
as whether normal or disease affected. During analysis, the results of the proposed MBO-MLPNN technique is
validated and compared by using various measures. Here the proposed method achieved 98% accuracy
performance for heart attack detection than former methods.