Deep diagnostic agent forest (DDAF): A deep learning pathogen recognition system for pneumonia based on CT. (February 2022)

• Record Type: Journal Article
• Authors: Chen, Weixiang; Han, Xiaoyu; Wang, Jian; Cao, Yukun; Jia, Xi; Zheng, Yuting; Zhou, Jie; Zeng, Wenjuan; Wang, Lin; Shi, Heshui; Feng, Jianjiang
• Is Part Of: Computers in biology and medicine. Volume 141(2022)
• Publication Date: 2022-02
• Subjects: Pathogens of pneumonia -- Deep learning -- Imbalanced data; AUC area under curves of receiver -- CAP community-acquired pneumonia -- CT computed tomography -- CXR Chest X-Ray -- DDAF deep diagnostic agent forest -- GBD Global Burden of Diseases -- HAP hospital-acquired pneumonia; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.105143
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Background: Even though antibiotics agents are widely used, pneumonia is still one of the most common causes of death around the world. Some severe, fast-spreading pneumonia can even cause huge influence on global economy and life security. In order to give optimal medication regimens and prevent infectious pneumonia's spreading, recognition of pathogens is important. Method: In this single-institution retrospective study, 2, 353 patients with their CT volumes are included, each of whom was infected by one of 12 known kinds of pathogens. We propose Deep Diagnostic Agent Forest (DDAF) to recognize the pathogen of a patient based on ones' CT volume, which is a challenging multiclass classification problem, with large intraclass variations and small interclass variations and very imbalanced data. Results: The model achieves 0.899 ± 0.004 multi-way area under curves of receiver (AUC) for level-I pathogen recognition, which are five rough groups of pathogens, and 0.851 ± 0.003 AUC for level-II recognition, which are 12 fine-level pathogens. The model also outperforms the average result of seven human readers in level-I recognition and outperforms all readers in level-II recognition, who can only reach an average result of 7.71 ± 4.10% accuracy. Conclusion: Deep learning model can help in recognition pathogens using CTs only, which might help accelerate the process of etiological diagnosis. Highlights: We proposed a deep learning system to recognize 12 types of pneumonia pathogens using only CTs. The system works with 0.851 ± 0.003 AUC in a 12-catogery database with 2353 cases and imbalanced distribution. The system outperforms the averaged of seven human radiologists in the reader studies.