CNN Based Multi-Feature Fusion with Metaheuristic Algorithms for Effective Feature Extraction and Classification OF 2D Echo Cardiovascular Diseases

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 3

Year of Publication : 2025

Authors : K. Deepthi Reddy, N. Pushpalatha, Venkata Ramana M., Pallapati Ravi Kumar, J. Manoranjini, E. Gurumoorthi, Puligilla Sridevi

10.14445/23488549/IJECE-V12I3P117

How to Cite?

K. Deepthi Reddy, N. Pushpalatha, Venkata Ramana M., Pallapati Ravi Kumar, J. Manoranjini, E. Gurumoorthi, Puligilla Sridevi, "CNN Based Multi-Feature Fusion with Metaheuristic Algorithms for Effective Feature Extraction and Classification OF 2D Echo Cardiovascular Diseases," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 3, pp. 171-178, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I3P117

Abstract:

Deep learning offers enormous potential to improve ultrasound quality through real-time heart anatomy and function analysis for clinical echocardiography and point-of-care diagnostics. Machine learning makes automating processes like echocardiography analysis, quality rating, view categorization, heart area segmentation, and diagnostic index computation easier. By extracting characteristics through data augmentation, existing approaches effectively categorize 2D echo data using high-performance deep neural networks. Using the Multi-Feature-Fusion (MFF) model, which combines wavelet packet energy, fuzzy entropy, and optimization algorithms for feature extraction, our system presents an innovative and efficient approach for analyzing and quantifying echocardiogram in real time. Using learned representations to improve target echo task learning, a Convolution Neural Network (CNN) has been trained on a large public dataset. The CNN integrates optimization techniques such as squirrel and crow meta-heuristics for efficient 2D echocardiography feature extraction, boundary identification, and image classification. A module locates regions of interest, and three thin routes extract high-level attributes and low-level texture. The model demonstrates its strong performance in reaching an accuracy of 98.2% for anomaly recognition, as evidenced by evaluation measures such as accuracy, specificity, sensitivity, precision, and AUC. This highlights the efficiency of our deep learning method, Multi-Feature Fusion, for the interpretation and quantification of Echocardiography in real-time.

Keywords:

Echocardiography, Deep Learning, Multi-feature-fusion model, Real-time analysis, Abnormality recognition.

References:

[1] Jae K. Oh, James B. Seward, and A. Jamil Tajik, The Echo Manual, Lippincott Williams & Wilkins, pp. 1-431, 2006.
[Google Scholar] [Publisher Link]
[2] Ghada Zamzmi et al., “Harnessing Machine Intelligence in Automatic Echocardiogram Analysis: Current Status, Limitations, and Future Directions,” IEEE Reviews in Biomedical Engineering, vol. 14, pp. 181-203, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Ghada Zamzmi, Sivaramakrishnan Rajaraman, and Sameer Antani, “UMS-Rep: Unified Modality-Specific Representation for Efficient Medical Image Analysis, “Informatics in Medicine Unlocked, vol. 24, pp. 1-12, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Lisa Torrey, and Jude Shavlik, Handbook of Research on Machine Learning Applications and Trends: Algorithms, Methods, and Techniques, Information Science Reference - Imprint of: IGI Publishing, pp. 242-264, 2010.
[Google Scholar] [Publisher Link]
[5] Karen Simonyan, and Andrew Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” arXiv Preprint, pp. 1-14, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Erik Smistad et al., “Real-Time Automatic Ejection Fraction and Foreshortening Detection Using Deep Learning,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 67, no. 12, pp. 2595-2604, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Chao Dong et al., “Learning a Deep Convolutional Network for Image Super-Resolution,” Proceedings 13th European Conference Computer Vision, Zurich, Switzerland, pp. 184-199, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Olga Russakovsky et al., “ImageNet Large Scale Visual Recognition Challenge,” International Journal of Computer Vision, vol. 115, pp. 211-252, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Christian Szegedy et al., “Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning Intelligence,” Proceedings of the AAAI Conference on Artificial Intelligence, San Francisco, California, USA, pp. 4278-4284, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Sergey Ioffe, and Christian Szegedy, “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate,” Proceedings of the 32nd International Conference on Machine Learning, vol. 37, pp. 448-456, 2015.
[Google Scholar] [Publisher Link]
[11] Christian Szegedy et al., “Rethinking the Inception Architecture for Computer Vision,” IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, pp. 2818-2826, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Kaiming He et al., “Deep Residual Learning for Image Recognition,” IEEE Conference on Computer Vision and Pattern Recognition,” Las Vegas, NV, USA, pp. 770-778, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Inayathullah Ghori et al., “Echocardiogram Analysis Using Motion Profile Modeling,” IEEE Transactions on Medical Imaging, vol. 39, no. 5, pp. 1767-1774, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Yuzhe Liu, Vanathi Gopalakrishnan, and Shobhit Madan, “Quantitative Clinical Guidelines for Imaging Use in Evaluation of Pediatric Cardiomyopathy,” IEEE International Conference on Bioinformatics and Biomedicine, Washington, DC, USA, pp. 1572-1578, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Amir H. Abdi et al., “Automatic Quality Assessment of Echocardiograms Using Convolutional Neural Networks: Feasibility on the Apical Four-Chamber View,” IEEE Transactions on Medical Imaging, vol. 36, no. 6, pp. 1221-1230 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Razvan O. Mada et al., “How to Define End-Diastole and End-Systole?,” JACC: Cardiovascular Imaging, vol. 8, no. 2, pp. 148-157, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Fatemeh Taheri Dezaki et al., “Cardiac Phase Detection in Echocardiograms with Densely Gated Recurrent Neural Networks and Global Extrema Loss,” IEEE Transactions on Medical Imaging, vol. 38, no. 8, pp. 1821-1832, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Gene R. Quinn et al., “Missed Diagnosis of Cardiovascular Disease in Outpatient General Medicine: Insights from Malpractice Claims Data,” The Joint Commission Journal on Quality and Patient Safety, vol. 43, no. 10, pp. 508-516, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Maria Panagioti et al., “Association between Physician Burnout and Patient Safety, Professionalism, and Patient Satisfaction: A systematic Review and Meta-analysis,” JAMA Internal Medicine, vol. 178, no. 10, pp. 1317-1331, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[20] U. Barcaro, D. Moroni, and O. Salvetti, “Automatic Computation of Left Ventricle Ejection Fraction from Dynamic Ultrasound Images,” Pattern Recognition and Image Analysis,” vol. 18, pp. 351-358, 2008.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Saeed Darvishi et al., “Measuring Left Ventricular Volumes in Two-Dimensional Echocardiogra-phy Image Sequence Using Level-set Method for Automatic Detection of End-Diastole and End-systole Frames,” Research in Cardiovascular Medicine, vol. 2, no. 1, pp. 39-45, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Eleonora Sulas et al., “Automatic Detection of Complete and Measurable Cardiac Cycles in Antenatal Pulsed-Wave Doppler Signals,” Computer Methods and Programs in Biomedicine, vol. 190, pp. 1-10, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Sarah Leclerc et al., “Deep Learning for Segmentation Using an Open Large-Scale Dataset in 2D Echocardiography,” IEEE Transactions on Medical Imaging, vol. 38, no. 9, pp. 2198-2210, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Yi Guo et al., “Left Ventricle Volume Measuring using Echocardiography Sequences,” 2018 Digital Image Computing: Techniques and Applications (DICTA), Canberra, ACT, pp. 1-8, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Malik Saad Sultan et al., “Virtual M-Mode for Echocardiography: A New Approach for the Segmentation of the Anterior Mitral Leaflet,” IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 1, pp. 305-313, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Vinícius Veloso de Melo et al., “Gradient Boosting Decision Trees for Echocardiogram Images,” International Joint Conference on Neural Networks, Rio de Janeiro, Brazil, pp. 1-8, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Yuanwei Li et al., “Fully Automatic Myocardial Segmentation of Contrast Echocardiography Sequence Using Random Forests Guided by Shape Model,” IEEE Transactions on Medical Imaging, vol. 37, no. 5, pp. 1081-1091, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Hanie Moghaddasi, and Saeed Nourian, “Automatic Assessment of Mitral Regurgitation Severity based on Extensive Textural Features on 2D Echocardiography Videos,” Computers in Biology and Medicine, vol. 73, pp. 47-55, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Ali Madani et al., “Deep Echocardiography: Data-Efficient Supervised and Semi-Supervised Deep Learning towards Automated Diagnosis of Cardiac Disease,” NPJ Digital Medicine, vol. 1, pp. 1-11, 2018.
[CrossRef] [Google Scholar] [Publisher Link]