Cited 0 times in
Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 김휘영 | - |
dc.date.accessioned | 2024-05-30T06:57:39Z | - |
dc.date.available | 2024-05-30T06:57:39Z | - |
dc.date.issued | 2023-07 | - |
dc.identifier.issn | 1738-3536 | - |
dc.identifier.uri | https://ir.ymlib.yonsei.ac.kr/handle/22282913/199477 | - |
dc.description.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 medica-tion, 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 perfor-mance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to de-termine active TB by DL model and radiologists. Sensitivities and specificities for deter- mining 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 val-ues 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 TB burden countries. | - |
dc.description.statementOfResponsibility | open | - |
dc.language | Korean | - |
dc.publisher | 대한결핵 및 호흡기학회 | - |
dc.relation.isPartOf | TUBERCULOSIS AND RESPIRATORY DISEASES | - |
dc.rights | CC BY-NC-ND 2.0 KR | - |
dc.title | Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm | - |
dc.type | Article | - |
dc.contributor.college | College of Medicine (의과대학) | - |
dc.contributor.department | Dept. of Biomedical Systems Informatics (의생명시스템정보학교실) | - |
dc.contributor.googleauthor | Ye Ra Choi | - |
dc.contributor.googleauthor | Soon Ho Yoon | - |
dc.contributor.googleauthor | Jihang Kim | - |
dc.contributor.googleauthor | Jin Young Yoo | - |
dc.contributor.googleauthor | Hwiyoung Kim | - |
dc.contributor.googleauthor | Kwang Nam Jin | - |
dc.identifier.doi | 10.4046/trd.2023.0020 | - |
dc.contributor.localId | A05971 | - |
dc.relation.journalcode | J02761 | - |
dc.identifier.eissn | 2005-6184 | - |
dc.identifier.pmid | 37183400 | - |
dc.subject.keyword | Artificial Intelligence | - |
dc.subject.keyword | Chest Radiography | - |
dc.subject.keyword | Deep Learning Algorithm | - |
dc.subject.keyword | Tuberculosis | - |
dc.contributor.alternativeName | Kim, Hwiyoung | - |
dc.contributor.affiliatedAuthor | 김휘영 | - |
dc.citation.volume | 86 | - |
dc.citation.number | 3 | - |
dc.citation.startPage | 226 | - |
dc.citation.endPage | 233 | - |
dc.identifier.bibliographicCitation | TUBERCULOSIS AND RESPIRATORY DISEASES, Vol.86(3) : 226-233, 2023-07 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.