Deep learning-based pulmonary tuberculosis automated detection on chest radiography: large-scale independent testing

Quantitative Imaging in Medicine and Surgery

article

Wen Zhou¹  Guanxun Cheng¹  Ziqi Zhang³  Litong Zhu⁴  Stefan Jaeger⁵  Fleming Y. M. Lure⁶  Lin Guo⁶ 

[1] Department of Radiology , Peking University Shenzhen Hospital;Department of Radiology , Peking University First Hospital;Tsinghua-Berkeley Shenzhen Institute , Tsinghua University;Department of Medicine, Queen Mary Hospital, University of Hong Kong;National Library of Medicine , National Institutes of Health;Shenzhen Smart Imaging Healthcare Co., Ltd.

关键词: Tuberculosis detection;    deep learning;    chest radiography;    external validation;    large-scale test;   

DOI  :  10.21037/qims-21-676

学科分类：外科医学

来源: AME Publications

【 摘 要 】

Background: It is critical to have a deep learning-based system validated on an external dataset before it is used to assist clinical prognoses. The aim of this study was to assess the performance of an artificial intelligence (AI) system to detect tuberculosis (TB) in a large-scale external dataset. Methods: An artificial, deep convolutional neural network (DCNN) was developed to differentiate TB from other common abnormalities of the lung on large-scale chest X-ray radiographs. An internal dataset with 7,025 images was used to develop the AI system, including images were from five sources in the U.S. and China, after which a 6-year dynamic cohort accumulation dataset with 358,169 images was used to conduct an independent external validation of the trained AI system. Results: The developed AI system provided a delineation of the boundaries of the lung region with a Dice coefficient of 0.958. It achieved an AUC of 0.99 and an accuracy of 0.948 on the internal data set, and an AUC of 0.95 and an accuracy of 0.931 on the external data set when it was used to detect TB from normal images. The AI system achieved an AUC of more than 0.9 on the internal data set, and an AUC of over 0.8 on the external data set when it was applied to detect TB, non-TB abnormal and normal images. Conclusions: We conducted a real-world independent validation, which showed that the trained system can be used as a TB screening tool to flag possible cases for rapid radiologic review and guide further examinations for radiologists.

【 授权许可】

All Rights reserved