| Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm |
|
Ye Ra Choi, MD1,2, Soon Ho Yoon, MD, PhD2,3, Jihang Kim, MD, PhD2,4, Jin Young Yoo, MD5, Hwiyoung Kim, PhD6, Kwang Nam Jin, MD, PhD1,2 |
1Department of Radiology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul, Republic of Korea
2Department of Radiology, Seoul National University College of Medicine, Seoul, Republic of Korea
3Department of Radiology, Seoul National University Hospital, Seoul, Republic of Korea
4Department of Radiology, Seoul National University Bundang Hospital, Seongnam, Republic of Korea
5Department of Radiology, Chungbuk National University Hospital, Cheongju, Republic of Korea
6Department of Radiology and Research Institute of Radiologic Science, Yonsei University College of Medicine, Seoul, Republic of Korea |
Correspondence:
Kwang Nam Jin, Tel: 82-2-870-2536, Fax: 82-2-831-2826,
Email: wlsrhkdska@gmail.com
|
|
Received: 22 February 2023 • Revised: 11 April 2023 • Accepted: 14 May 2023 |
| Abstract |
Background
Inactive or old, healed tuberculosis (TB) on chest radiograph (CR) is often found in high TB incidence countries, and to avoid unnecessary evaluation and medication, differentiation from active TB is important. This study develops a deep learning (DL) model to estimate activity in a single chest radiographic analysis.
Methods
A total of 3,824 active TB CRs from 511 individuals and 2,277 inactive TB CRs from 558 individuals were retrospectively collected. A pretrained convolutional neural network was fine-tuned to classify active and inactive TB. The model was pretrained with 8,964 pneumonia and 8,525 normal cases from the National Institute of Health (NIH) dataset. During the pretraining phase, the DL model learns the following tasks: pneumonia vs. normal, pneumonia vs. active TB, and active TB vs. normal. The performance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to determine active TB by DL model and radiologists. Sensitivities and specificities for determining active TB were evaluated for both the DL model and radiologists.
Results
The performance of the DL model showed area under the curve (AUC) values of 0.980 in internal validation, and 0.815 and 0.887 in external validation. The AUC values for the DL model, thoracic radiologist, and general radiologist, evaluated using one of the external validation datasets, were 0.815, 0.871, and 0.811, respectively.
Conclusion
This DL-based algorithm showed potential as an effective diagnostic tool to identify TB activity, and could be useful for the follow-up of patients with inactive TB in high TB burden countries. |
| Key Words:
Chest Radiography, Tuberculosis, Artificial Intelligence, Deep Learning Algorithm |
|
PDF Links
Data Sharing Statement
Full text via DOI
CrossRef TDM
Download Citation
