(Systems and Biomedical Engineering, Cairo University, Egypt)(2) Neveen Abd-Elsalam
(Systems and Biomedical Engineering, Cairo University, Egypt)(3) Ahmed Khandil
(Systems and Biomedical Engineering, Cairo University, Egypt)(4) Ahmed Elbialy
(Systems and Biomedical Engineering, Cairo University; and Shorouk Academy, Egypt)(5) Abou-Bakr Youssef
(Systems and Biomedical Engineering, Cairo University, Egypt)*corresponding author
AbstractTo advance the automated interpretation of lung ultrasound (LUS) data, multiple deep learning (DL) models have been introduced to identify LUS patterns for differentiating COVID-19, Pneumonia, and Normal cases. While these models have generally yielded promising outcomes, they have encountered challenges in accurately classifying each pattern across diverse cases. Therefore, this study introduces an ensemble framework that leverages multiple classification models, optimizing their contributions to the final prediction through a majority voting mechanism. After training seven different classification models, the three models with the highest accuracies were selected. The ensemble incorporates these top-performing models: EfficientNetV2-B0, EfficientNetV2-B2, and EfficientNetV2-B3, and utilizes this framework to classify patterns in LUS images. Compared to individual model performance, the ensemble approach significantly enhances classification accuracy, achieving an accuracy of 99.25% and an F1-score of 99%. In contrast, the standalone models attained accuracies of 97.8%, 97.6%, and 98.1%, with F1-score of approximately 98%. This research highlights the potential of ensemble learning for improving the accuracy and robustness of automated LUS analysis, offering a practical and scalable solution for real-world medical diagnostics. By combining the strengths of multiple models, the proposed framework paves the way for more reliable and efficient tools to assist clinicians in diagnosing lung diseases.
KeywordsCOVID-19; Pneumonia; Deep Learning; Transfer Learning; Ensemble method
|
DOIhttps://doi.org/10.26555/ijain.v11i1.1966 |
Article metricsAbstract views : 844 | PDF views : 136 |
Cite |
Full Text Download
|
References
[1] J. L. Díaz-Gómez, P. H. Mayo, and S. J. Koenig, “Point-of-Care Ultrasonography,” N. Engl. J. Med., vol. 385, no. 17, pp. 1593–1602, Oct. 2021, doi: 10.1056/NEJMra1916062.
[2] H. Scott, A. Zahra, R. Fernandes, B. C. Fries, H. C. Thode, and A. J. Singer, “Bacterial infections and death among patients with Covid-19 versus non Covid-19 patients with pneumonia,” Am. J. Emerg. Med., vol. 51, pp. 1–5, Jan. 2022, doi: 10.1016/j.ajem.2021.09.040.
[3] P. R. Joshi, “Pulmonary Diseases in Older Patients: Understanding and Addressing the Challenges,” Geriatrics, vol. 9, no. 2, p. 34, Mar. 2024, doi: 10.3390/geriatrics9020034.
[4] D. Davis et al., “Advancements in the Management of Severe Community-Acquired Pneumonia: A Comprehensive Narrative Review,” Cureus, vol. 15, no. 10, Oct, pp. 1-11, 2023, doi: 10.7759/cureus.46893.
[5] World Health Organization “Coronavirus disease (COVID-19).” [Online]. Available at: https://www.who.int/emergencies/diseases/novel-coronavirus-2019.
[6] F. J. Rodríguez-Contreras et al., “Lung Ultrasound Performed by Primary Care Physicians for Clinically Suspected Community-Acquired Pneumonia: A Multicenter Prospective Study,” Ann. Fam. Med., vol. 20, no. 3, pp. 227–236, May 2022, doi: 10.1370/afm.2796.
[7] J. Born et al., “POCOVID-Net: Automatic Detection of COVID-19 From a New Lung Ultrasound Imaging Dataset (POCUS),” Eur. PMC, no. April, pp. 1–14, 2020. [Online]. Available at: https://europepmc.org/article/PPR/PPR268507.
[8] S. Bahri, Suprijanto, and E. Juliastuti, “Texture Analysis of Ultrasound Images to Differentiate Pneumonia and Covid-19,” in IEEE International Biomedical Instrumentation and Technology Conference (IBITeC), Yogyakarta, Indonesia, 2021, pp. 24–28, doi: 10.1109/IBITeC53045.2021.9649067.
[9] M. La Salvia et al., “Deep learning and lung ultrasound for Covid-19 pneumonia detection and severity classification,” Comput. Biol. Med., vol. 136, p. 104742, Sep. 2021, doi: 10.1016/j.compbiomed.2021.104742.
[10] O. Elkhuoly, M. Malhat, A. Keshk, and M. Elsabaawy, “A Comparative Analysis of COVID-19 Diagnosis Using Lung Ultrasound Based on Convolutional Neural Networks,” IJCI. Int. J. Comput. Inf., vol. 10, no. 1, pp. 0–0, Sep. 2022, doi: 10.21608/ijci.2022.151629.1079.
[11] S. Shang et al., “Performance of a computer aided diagnosis system for SARS-CoV-2 pneumonia based on ultrasound images,” Eur. J. Radiol., vol. 146, p. 110066, Jan, doi: 10.1016/j.ejrad.2021.110066.
[12] S. Morsy et al., “Impact of Dataset Size on the Performance of Deep Learning Models: A Case Study on Lung Ultrasound and X-Ray Image Classification,” SSRN Electron. J., vol. 8, no. 4, pp. 21–30, Aug. 2024, doi: 10.2139/ssrn.5044890.
[13] E. A. Nehary, S. Rajan, and C. Rossa, “Lung Ultrasound Image Classification Using Deep Learning and Histogram of Oriented Gradients Features for COVID-19 Detection,” in 2023 IEEE Sensors Applications Symposium (SAS), 2023, pp. 1–6, doi: 10.1109/SAS58821.2023.10254002.
[14] J. Himasree, K. Aravindhan, K. P. Keerthana, and C. Gobinath, “Design of an Efficient Deep Learning Framework for Covid-19 Image Classification,” in 2023 International Conference on New Frontiers in Communication, Automation, Management and Security (ICCAMS), 2023, vol. 1, pp. 1–4, doi: 10.1109/ICCAMS60113.2023.10525806.
[15] A. Subramanyam and M. Sucharitha, “Multi-classification of Lung Diseases Using Lung Ultrasound Imaging BT - AI Technologies for Information Systems and Management Science,” 2024, pp. 510–521, doi: 10.1007/978-3-031-66410-6_40.
[16] G. Madhu, S. Kautish, Y. Gupta, G. Nagachandrika, S. M. Biju, and M. Kumar, “XCovNet: An optimized xception convolutional neural network for classification of COVID-19 from point-of-care lung ultrasound images,” Multimed. Tools Appl., vol. 83, no. 11, pp. 33653–33674, 2024, doi: 10.1007/s11042-023-16944-z.
[17] P. T. H. Nhat et al., “Clinical benefit of AI-assisted lung ultrasound in a resource-limited intensive care unit.,” Crit. Care, vol. 27, no. 1, p. 257, Jul. 2023, doi: 10.1186/s13054-023-04548-w.
[18] L. Howell, N. Ingram, R. Lapham, A. Morrell, and J. R. McLaughlan, “Deep learning for real-time multi-class segmentation of artefacts in lung ultrasound.,” Ultrasonics, vol. 140, p. 107251, May 2024, doi: 10.1016/j.ultras.2024.107251.
[19] W. Xing et al., “Automated lung ultrasound scoring for evaluation of coronavirus disease 2019 pneumonia using two-stage cascaded deep learning model,” Biomed. Signal Process. Control, vol. 75, p. 103561, May 2022, doi: 10.1016/j.bspc.2022.103561.
[20] R. Chaudhary et al., “Diagnostic accuracy of an automated classifier for the detection of pleural effusions in patients undergoing lung ultrasound,” Am. J. Emerg. Med., vol. 90, pp. 142–150, Apr. 2025, doi: 10.1016/j.ajem.2025.01.041.
[21] A. K. Dubey et al., “Ensemble Deep Learning Derived from Transfer Learning for Classification of COVID-19 Patients on Hybrid Deep-Learning-Based Lung Segmentation: A Data Augmentation and Balancing Framework,” Diagnostics, vol. 13, no. 11, p. 1954, Jun. 2023, doi: 10.3390/diagnostics13111954.
[22] D. E. Shea et al., “Deep Learning Video Classification of Lung Ultrasound Features Associated with Pneumonia,” in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2023, pp. 3103–3112, doi: 10.1109/CVPRW59228.2023.00312.
[23] U. Khan, A. Smargiassi, R. Inchingolo, and L. Demi, “A Novel Weighted Majority Voting-Based Ensemble Framework for Lung Ultrasound Pattern Classification in Pneumonia Patients,” in 2023 IEEE International Ultrasonics Symposium (IUS), 2023, pp. 1–4, doi: 10.1109/IUS51837.2023.10308194.
[24] M. Iman, K. Rasheed, and H. R. Arabnia, “A Review of Deep Transfer Learning and Recent Advancements,” pp. 1-14, Jan. 2022, doi: 10.3390/technologies11020040.
[25] X. Li et al., “Principled and efficient transfer learning of deep models via neural collapse,” arXiv Prepr. arXiv2212.12206, pp. 1-30, 2022. [Online]. Available at: https://arxiv.org/abs/2212.12206.
[26] M. W. Ahdi, Khalid, A. Kunaefi, B. A. Nugroho, and A. Yusuf, “Convolutional Neural Network (CNN) EfficientNet-B0 Model Architecture for Paddy Diseases Classification,” in 2023 14th International Conference on Information & Communication Technology and System (ICTS), Oct. 2023, pp. 105–110, doi: 10.1109/ICTS58770.2023.10330828.
[27] M. Tan and Q. Le, “Efficientnet: Rethinking model scaling for convolutional neural networks,” in International conference on machine learning, 2019, pp. 6105–6114. [Online]. Available at: https://arxiv.org/abs/1905.11946v5.
[28] M. Tan and Q. V. Le, “EfficientNetV2: Smaller Models and Faster Training,” in Proceedings of Machine Learning Research, 2021, vol. 139, pp. 10096–10106, [Online]. Available at: https://arxiv.org/abs/1905.11946.
[29] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016, vol. 2016-Decem, pp. 2818–2826, doi: 10.1109/CVPR.2016.308.
[30] A. Arini, M. Azhari, I. I. A. Fitri, and F. Fahrianto, “Performance Analysis of Transfer Learning Models for Identifying AI-Generated and Real Images,” J. Tek. Inform., vol. 17, no. 2, pp. 139–152, Oct. 2024, doi: 10.15408/jti.v17i2.40453.
[31] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016, vol. 2016-Decem, pp. 770–778, doi: 10.1109/CVPR.2016.90.
[32] N. Parimala and G. Muneeswari, “Modified ResNet152v2: Binary Classification and Hybrid Segmentation of Brain Stroke Using Transfer Learning-Based Approach,” Polish J. Med. Phys. Eng., vol. 30, no. 1, pp. 24–35, Mar. 2024, doi: 10.2478/pjmpe-2024-0004.
[33] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. Alemi, “Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning,” Proc. AAAI Conf. Artif. Intell., vol. 31, no. 1, pp. 4278–4284, Feb. 2017, doi: 10.1609/aaai.v31i1.11231.
[34] R. P. Tripathi, S. K. Khatri, and D. V. G. Baxodirovna, “A transfer learning approach to implementation of pretrained CNN models for Breast cancer diagnosis,” J. Posit. Sch. Psychol., pp. 5816–5830, 2022. [Online]. Available at: https://journalppw.com/index.php/jpsp/article/view/4358.
[35] P. Schneider and F. Xhafa, Anomaly Detection and Complex Event Processing Over IoT Data Streams: With Application to EHealth and Patient Data Monitoring. Elsevier Science, pp. 1-406, 2022, doi: 10.1016/B978-0-12-823818-9.00013-4.
[36] I. D. Mienye and Y. Sun, “A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects,” IEEE Access, vol. 10, pp. 99129–99149, 2022, doi: 10.1109/ACCESS.2022.3207287.
[37] M. A. Ganaie, M. Hu, A. K. Malik, M. Tanveer, and P. N. Suganthan, “Ensemble deep learning: A review,” Eng. Appl. Artif. Intell., vol. 115, p. 105151, 2022, doi: 10.1016/j.engappai.2022.105151.
[38] A. Chatzimparmpas, R. M. Martins, K. Kucher, and A. Kerren, “StackGenVis: Alignment of data, algorithms, and models for stacking ensemble learning using performance metrics,” IEEE Trans. Vis. Comput. Graph., vol. 27, no. 2, pp. 1547–1557, Feb. 2021, doi: 10.1109/TVCG.2020.3030352.
[39] A. Desiani et al., “Majority Voting as Ensemble Classifier for Cervical Cancer Classification,” Sci. Technol. Indones., vol. 8, no. 1 SE-Articles, pp. 84–92, Jan. 2023, doi: 10.26554/sti.2023.8.1.84-92.
[40] M. Azad, T. H. Nehal, and M. Moshkov, “A novel ensemble learning method using majority based voting of multiple selective decision trees,” Computing, vol. 107, no. 1, p. 42, 2024, doi: 10.1007/s00607-024-01394-8.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0