Rima Ayu Cahyani, Aji Purwinarko

Neural Network Optimization Using Hybrid Adaptive Mutation Particle Swarm Optimization and Levenberg-Marquardt in Cases of Cardiovascular Disease

Introduction

Neural network optimization using hybrid adaptive mutation particle swarm optimization and levenberg-marquardt in cases of cardiovascular disease. Optimize neural networks for cardiovascular disease prediction using Hybrid AMPSO-LMA. This study improves MLP accuracy from 86.1% to 86.88% for early CVD detection, reducing mortality.

Abstract

Abstract. Cardiovascular disease is a condition generally characterized by the narrowing or blockage of blood vessels, which can lead to heart attacks, chest pain, or strokes. It is the leading cause of death worldwide, accounting for approximately 31% or 17.9 million deaths each year globally. Deaths caused by cardiovascular disease are projected to continue increasing until 2030, with the number of patients reaching 23.3 million. As cases of death due to cardiovascular disease become more prevalent, early detection is crucial to reduce mortality rates. Purpose: Many previous researchers have conducted studies on predicting cardiovascular disease using neural network methods. This study extends these methods by incorporating feature selection and optimization with Hybrid AMPSO-LMA. The research is designed to explore the implementation and predictive outcomes of Hybrid AMPSO-LMA in optimizing MLP for cases of cardiovascular disease. Methods/Study design/approach: The first step in conducting this research is to download the Heart Disease Dataset from Kaggle.com. The dataset is processed through preprocessing by removing duplicates and transforming the data. Then, data mining processes are carried out using the MLP algorithm optimized with Hybrid AMPSO-LMA to obtain results and conclusions. This system is designed using the Python programming language and utilizes Flask for website access in HTML. Result/Findings: The research results demonstrate that the method employed by the author successfully improves the accuracy of predicting cardiovascular disease. Predicting cardiovascular disease using the MLP algorithm yields an accuracy of 86.1%, and after optimization with Hybrid AMPSO-LMA, the accuracy increases to 86.88%. Novelty/Originality/Value: This effort will contribute to the development of a more reliable and effective cardiovascular disease prediction system, with the goal of early identification of individuals exhibiting symptoms of cardiovascular disease.

Review

This paper presents a method for optimizing neural network performance in the critical area of cardiovascular disease (CVD) prediction, a leading global cause of mortality. The authors propose integrating a novel Hybrid Adaptive Mutation Particle Swarm Optimization (AMPSO) and Levenberg-Marquardt Algorithm (LMA), termed Hybrid AMPSO-LMA, to enhance the accuracy of Multi-Layer Perceptron (MLP) models. Using a Kaggle Heart Disease Dataset, the study's primary objective is to demonstrate the practical application and improved predictive outcomes of this hybrid optimization strategy. The initial results indicate a positive, albeit modest, increase in prediction accuracy following the application of their proposed method. While addressing a highly relevant medical challenge, the abstract highlights several areas for clarification and further depth. The claim of incorporating "feature selection" is mentioned in the purpose but is not elaborated upon in the methods or results sections, which primarily focus on network optimization. This leaves ambiguity regarding its actual implementation and contribution. Furthermore, the reported accuracy improvement from 86.1% to 86.88% is numerically small. A robust analysis would benefit from discussing the statistical significance of this improvement and its practical implications in a clinical context, rather than solely presenting the numerical gain. A more detailed explanation of the specific hybridization mechanism within "Hybrid AMPSO-LMA" would also be valuable for understanding the methodological novelty. Despite these observations, the core objective of developing a more reliable system for early CVD detection remains highly significant. To build upon this effort, future work should consider a more extensive comparative analysis, benchmarking the proposed Hybrid AMPSO-LMA against a wider range of state-of-the-art machine learning models and optimization techniques prevalent in medical diagnostics. Evaluating the method's performance on larger, more diverse, or clinically validated datasets would also be crucial to assess its generalizability and real-world applicability. Additionally, exploring the interpretability of the optimized network, particularly if feature selection is indeed integral, could provide valuable insights into the key predictors of cardiovascular disease.

Full Text

Comments

You need to be logged in to post a comment.