volume | PIER Journals

Abstract

Severe bacterial pneumonia is a serious respiratory disease caused by bacteria, which is mainly transmitted through the respiratory tract. To achieve early recognition of severe pneumonia patients through images, this study collected the CT images of 180 patients diagnosed with bacterial infection in the lungs on the day of emergency admission to a large regional medical center (a Top-Tier (Grade 3 A) hospital). After classification by two deputy chief physicians of the respiratory department, 93 cases of severe bacterial infection were obtained and the rest 87 cases were identified as mild bacterial infection. The CT sequences were then preprocessed and annotated to obtain 599 images with annotated lung infection areas. Together with 107 normal (non-infected) images, these bacterial infection images were randomly divided into a training set of 447 and a test set of 259. In the experiment, four deep learning methods, namely, FCN, PSPNet, deeplabv3, and deeplabv3plus, were used for training and three-class classification (severe bacterial infection, mild bacterial infection, and normal). Deeplabv3plus showed the best performance, with an overall accuracy of 96.91% (including a sensitivity of 95.25%, a specificity of 97.24%, an accuracy of 86.96%, a recall rate of 95.24%, and an F1 score of 90.91%) for severe bacterial infection. Using deep learning technology to diagnose severe pneumonia as early as possible can produce valuable treatment time for patients, thereby significantly reducing mortality and complication rates.

1. Jain, Seema, Wesley H. Self, Richard G. Wunderink, Sherene Fakhran, Robert Balk, Anna M. Bramley, Carrie Reed, Carlos G. Grijalva, Evan J. Anderson, D. Mark Courtney, et al. "Community-acquired pneumonia requiring hospitalization among U.S. adults," New England Journal of Medicine, Vol. 373, No. 5, 415-427, 2015.
doi:10.1056/NEJMoa1500245 Google Scholar

2. Vaughn, Valerie M., Robert P. Dickson, Jennifer K. Horowitz, and Scott A. Flanders, "Community-acquired pneumonia: A review," Jama, Vol. 332, No. 15, 1282-1295, 2024.
doi:10.1001/jama.2024.14796 Google Scholar

3. Cillóniz, Catiaa, Cristinab Dominedò, Carolinac Garcia-Vidal, and Antonia Torres, "Community-acquired pneumonia as an emergency condition," Current Opinion in Critical Care, Vol. 24, No. 6, 531-539, 2018.
doi:10.1097/MCC.0000000000000550 Google Scholar

4. Woodhead, Mark, Catherine A. Welch, David A. Harrison, Geoff Bellingan, and Jon G. Ayres, "Community-acquired pneumonia on the intensive care unit: Secondary analysis of 17,869 cases in the ICNARC Case Mix Programme Database," Critical Care, Vol. 10, S1, 2006.
doi:10.1186/cc4927 Google Scholar

5. Marti, Christophe, Nicolas Garin, Olivier Grosgurin, Antoine Poncet, Christophe Combescure, Sebastian Carballo, and Arnaud Perrier, "Prediction of severe community-acquired pneumonia: A systematic review and meta-analysis," Critical Care, Vol. 16, No. 4, R141, Jul. 2012.
doi:10.1186/cc11447 Google Scholar

6. Luo, Ning, Han Zhang, Yang Zhou, ZiXuan Kong, WeiHang Sun, Nan Huang, and AoDan Zhang, "Utility of chest CT in diagnosis of COVID-19 pneumonia," Diagnostic and Interventional Radiology, Vol. 26, No. 5, 437-442, 2020.
doi:10.5152/dir.2020.20144 Google Scholar

7. Kwon, TaeWoo, Sang Pyo Lee, Dongmin Kim, Jinseong Jang, Myungjae Lee, Shin Uk Kang, Heejin Kim, Keunyoung Oh, Jinhee On, Young Jae Kim, et al. "Diagnostic performance of artificial intelligence model for pneumonia from chest radiography," PLoS One, Vol. 16, No. 4, e0249399, Apr. 2021.
doi:10.1371/journal.pone.0249399 Google Scholar

8. Xu, Zhanpeng, Yiming Jiang, Jiali Ji, Erik Forsberg, Yuanpeng Li, and Sailing He, "Classification, identification, and growth stage estimation of microalgae based on transmission hyperspectral microscopic imaging and machine learning," Optics Express, Vol. 28, No. 21, 30686-30700, 2020.
doi:10.1364/oe.406036 Google Scholar

9. Zhu, He, Jing Luo, Jiaqi Liao, and Sailing He, "High-accuracy rapid identification and classification of mixed bacteria using hyperspectral transmission microscopic imaging and machine learning," Progress In Electromagnetics Research, Vol. 178, 49-62, 2023.
doi:10.2528/pier23082303 Google Scholar

10. Wang, Lei and Sailing He, "A 3-band iteration method to transfer knowledge learned in RGB pretrained models to hyperspectral domain," Progress In Electromagnetics Research M, Vol. 128, 1-9, 2024.
doi:10.2528/pierm23112103 Google Scholar

11. Zhang, Yu-han, Xiao-fei Hu, Jie-chao Ma, Xian-qi Wang, Hao-ran Luo, Zi-feng Wu, Shu Zhang, De-jun Shi, Yi-zhou Yu, Xiao-ming Qiu, et al. "Clinical applicable AI system based on deep learning algorithm for differentiation of pulmonary infectious disease," Frontiers in Medicine, Vol. 8, 753055, Dec. 2021.
doi:10.3389/fmed.2021.753055 Google Scholar

12. Huang, Tinglong, Xuelan Zheng, Lisui He, and Zhiliang Chen, "Diagnostic value of deep learning-based CT feature for severe pulmonary infection," Journal of Healthcare Engineering, Vol. 2021, No. 1, 5359084, Nov. 2021.
doi:10.1155/2021/5359084 Google Scholar

13. Long, Jonathan, Evan Shelhamer, and Trevor Darrell, "Fully convolutional networks for semantic segmentation," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 3431-3440, 2015.
doi:10.1109/cvpr.2015.7298965

14. Zhao, Hengshuang, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia, "Pyramid scene parsing network," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2881-2890, 2017.
doi:10.1109/cvpr.2017.660

15. Chen, Liang-Chieh, George Papandreou, Florian Schroff, and Hartwig Adam, "Rethinking atrous convolution for semantic image segmentation," Arxiv Preprint ArXiv:1706.05587, 2017.
doi:10.48550/arXiv.1706.05587 Google Scholar

16. Chen, Liang-Chieh, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam, "Encoder-decoder with atrous separable convolution for semantic image segmentation," Proceedings of the European Conference on Computer Vision (ECCV), 801-818, 2018.
doi:10.1007/978-3-030-01234-2_49

17. MMSegmentation Contributors "OpenMMLab semantic segmentation toolbox and benchmark," https://github.com/open-mmlab/mmsegmentation, 2020.

18. Hiremath, Amogh, Vidya Sankar Viswanathan, Kaustav Bera, Rakesh Shiradkar, Lei Yuan, Keith Armitage, Robert Gilkeson, Mengyao Ji, Pingfu Fu, Amit Gupta, et al. "Deep learning reveals lung shape differences on baseline chest CT between mild and severe COVID-19: A multi-site retrospective study," Computers in Biology and Medicine, Vol. 177, 108643, 2024.
doi:10.1016/j.compbiomed.2024.108643 Google Scholar

Dr. Ke Cui

Dr. Dawei Gong

Dr. Xiaobo Chen

Dr. Youzu Xu

Dr. Haiyan Li

Dr. Yefei Zhu

Dr. Julian Evans

Dr. Xin Gong

Dr. Zhenzhan Shi

Dr. Yinghe Xu

Prof. Sailing He