Efficient approach to predict the accuracy of heart disease by generating heartbeat based audio signal using resnet-50 compared with particle swarm optimization classifier

This study aims to assess the performance of Novel Resnet-50 compared to PSO (Particle Swarm Optimization) in predicting heart disease using audio signal-based heartbeat generation, with the goal of achieving higher accuracy. The innovative ResNet-50 and PSO models are used with a sample size of N=1...

Full description

Saved in:
Bibliographic Details
Main Authors: Muthaiah, M., Rekha, K. S., Narendran, R., Monika, E.
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Alpha rays
Confidence intervals
Design of experiments
Heart diseases
Particle swarm optimization
Performance prediction
Signal classification
Signal generation
Statistical analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study aims to assess the performance of Novel Resnet-50 compared to PSO (Particle Swarm Optimization) in predicting heart disease using audio signal-based heartbeat generation, with the goal of achieving higher accuracy. The innovative ResNet-50 and PSO models are used with a sample size of N=10. The experimental design specifies a pretest Gpower of 0.8, a significance level (alpha) of 0.05, and a confidence interval of 0.95. In terms of accuracy, the ResNet-50 model (96.40%) surpasses the PSO classifier (89.74%), as shown by a statistically significant p- value of 0.38. The statistical analysis utilizing the independent sample T-test demonstrates that there is no statistically significant difference between the two groups. The use of Novel ResNet-50 appears to improve the accuracy of the PSO classifier in this investigation.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0229210