A Hybrid Deep Learning Approach with Segmentation to Assess the Mitral Valve Failure using Ultrasound Images

DOI

10.2991/978-94-6463-754-0_14How to use a DOI?

Keywords

Mitral valve; classification; Accuracy; Ensemble Model; Deep Learning Networks

Abstract

Mitral valve abnormalities accurate detection is vital for diagnosing cardiovascular diseases. Most of the existing models which includes ResNet and U-Net has shown the exemplary performance in image segmentation process. But still there is a huge requirement in obtaining high precision and recall needed for reliable clinical implementation. We proposed an advanced hybrid model with the combined architecture of CNN, RNN, and R-CNN. The proposed architecture is designed specifically for enhancing the prediction on the presence of mitral valve conditions. The proposed hybrid model efficiency is compared with the existing approaches on various aspects. In particular, the proposed model has obtained an enhanced result over the existing approaches achieving pixel accuracy of 98.9%, mean Accuracy of 98.02%, Precision of 98.31%, recall of 97.56%, and a Dice Coefficient of 98.21%. The classification metrics were evaluated for various deep learning framework in which VGG16, DenseNet121, and ResNet152 combined as a fusion gave a high accuracy of 96%. This obtained result demonstrate that the hybrid technique enhances mitral valve detection accuracy and reliability, providing a reliable clinical solution.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - P. Sudha
AU  - R. Thirumalai Selvi
PY  - 2025
DA  - 2025/06/30
TI  - A Hybrid Deep Learning Approach with Segmentation to Assess the Mitral Valve Failure using Ultrasound Images
BT  - Proceedings of the 2025 International Conference on Advanced Research in Electronics and Communication Systems (ICARECS-2025)
PB  - Atlantis Press
SP  - 146
EP  - 161
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-754-0_14
DO  - 10.2991/978-94-6463-754-0_14
ID  - Sudha2025
ER  -