Ensemble Deep Learning Models for Heart Failure Prediction from Real Healthcare Records
- Version
- Download 60
- File Size 944.25 KB
- File Count 1
- Create Date 16 May 2024
- Last Updated 16 May 2024
Ensemble Deep Learning Models for Heart Failure Prediction from Real Healthcare Records
BG Sanchita1, Druthi K N2, Dr. Jagruthi H3
1 Student, B.E [Information Science and Engineering], BNM Institute of Technology, Bangalore
2 Student, B.E [Information Science and Engineering], BNM Institute of Technology, Bangalore
3 Associate Professor, Dept. of Information Science and Engineering, BNM Institute of Technology, Bangalore
Abstract— Cardiovascular diseases (CVDs) remain the
leading cause of death globally, necessitating advancements in non-invasive and efficient monitoring techniques. "Retinabeat" proposes a novel approach to CVD monitoring by leveraging the diagnostic potential of retinal vascular patterns. Utilizing state-of-the-art retinal imaging combined with the VGG19 convolutional neural network, this system aims to detect and analyze subtle changes in the retinal vasculature that correlate with cardiovascular health. Preliminary studies suggest that these retinal indicators can serve as reliable markers of cardiovascular anomalies, providing a non-invasive, cost-effective, and readily accessible method for early detection and ongoing management of heart health. This paper details the development and validation of the Retinabeat system, evaluates its performance against traditional methods, and discusses the implications of integrating advanced image processing technologies in routine cardiovascular care. The outcomes highlight the system’s potential to enhance predictive diagnostics and personalize health monitoring, paving the way for broader applications in telemedicine and preventive healthcare.
Keywords— Cardiovascular diseases, VGG19, telemedicine, Retinal images
Download