Detection of other pathologies when utilising computer-assisted digital solutions for TB screening
BACKGROUND: Computer-aided detection (CAD) tools for TB detection have the potential to enable screening programmes and reduce the diagnostic gap in settings where access to radiologists is limited. However, there are concerns that other common chest X-ray (CXR) abnormalities not due to TB may be mi...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
International Union Against Tuberculosis and Lung Disease (The Union)
2024-12-01
|
| Series: | IJTLD Open |
| Subjects: | |
| Online Access: | https://www.ingentaconnect.com/contentone/iuatld/ijtldo/2024/00000001/00000012/art00002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590523165048832 |
|---|---|
| author | J. Sebastian I.D. Olaru A. Giannakis M. Arentz S.V. Kik M. Ruhwald S. Linsen G. Günther P. Wolf F.J. Herth T. Weber C.M. Denkinger |
| author_facet | J. Sebastian I.D. Olaru A. Giannakis M. Arentz S.V. Kik M. Ruhwald S. Linsen G. Günther P. Wolf F.J. Herth T. Weber C.M. Denkinger |
| author_sort | J. Sebastian |
| collection | DOAJ |
| description | BACKGROUND: Computer-aided detection (CAD) tools for TB detection have the potential to enable screening programmes and reduce the diagnostic gap in settings where access to radiologists is limited. However, there are concerns that other common chest X-ray (CXR) abnormalities not due to TB may be missed. METHODS: We assessed the performance of three commercialised CAD tools (qXR, INSIGHT CXR and DrAID TM TB XR) to detect common non-TB abnormalities against readings with a standardised annotation guide by an expert radiologist. More than 20 well-characterised diagnoses besides TB significant in TB high-burden countries were examined. RESULTS: The 517 CXRs included were deemed abnormal by the three CAD with a sensitivity of respectively 97% (95% CI 95–98), 94% (95% CI 91–95), and 87% (95% CI 84–90) for INSIGHT CXR, qXR, and DrAID. The CAD generally detected abnormalities in patients with critical diagnoses such as lung cancer or heart failure. Performance for detecting other abnormalities was variable. CONCLUSION: This study showed that the three CAD tools identified CXRs as abnormal when diseases other than TB were present. Our findings alleviate ethical concerns of missing abnormalities other than TB when using commercially available CAD for TB screening and show their potential broader applicability. |
| format | Article |
| id | doaj-art-2a1e9f7e5b9544b1a9a264cb59ea3d0c |
| institution | Kabale University |
| issn | 3005-7590 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | International Union Against Tuberculosis and Lung Disease (The Union) |
| record_format | Article |
| series | IJTLD Open |
| spelling | doaj-art-2a1e9f7e5b9544b1a9a264cb59ea3d0c2025-01-23T13:29:23ZengInternational Union Against Tuberculosis and Lung Disease (The Union)IJTLD Open3005-75902024-12-0111253353910.5588/ijtldopen.24.04282Detection of other pathologies when utilising computer-assisted digital solutions for TB screeningJ. Sebastian0I.D. Olaru1A. Giannakis2M. Arentz3S.V. Kik4M. Ruhwald5S. Linsen6G. Günther7P. Wolf8F.J. Herth9T. Weber10C.M. Denkinger11Division of Infectious Disease and Tropical Medicine, Heidelberg University Hospital, Heidelberg, Germany;Division of Infectious Disease and Tropical Medicine, Heidelberg University Hospital, Heidelberg, Germany;Diagnostic and Interventional Radiology, Heidelberg University Hospital, Heidelberg, Germany;FIND, Geneva, Switzerland;FIND, Geneva, Switzerland;FIND, Geneva, Switzerland;FIND, Geneva, Switzerland;Department of Pulmonology and Allergology, Inselspital, Bern University Hospital, Bern, Switzerland;Department for Pneumology and Critical Care Medicine, Thoraxklinik, University of Heidelberg, Heidelberg, Germany;Department for Pneumology and Critical Care Medicine, Thoraxklinik, University of Heidelberg, Heidelberg, Germany;Diagnostic and Interventional Radiology, Heidelberg University Hospital, Heidelberg, Germany;Division of Infectious Disease and Tropical Medicine, Heidelberg University Hospital, Heidelberg, Germany;BACKGROUND: Computer-aided detection (CAD) tools for TB detection have the potential to enable screening programmes and reduce the diagnostic gap in settings where access to radiologists is limited. However, there are concerns that other common chest X-ray (CXR) abnormalities not due to TB may be missed. METHODS: We assessed the performance of three commercialised CAD tools (qXR, INSIGHT CXR and DrAID TM TB XR) to detect common non-TB abnormalities against readings with a standardised annotation guide by an expert radiologist. More than 20 well-characterised diagnoses besides TB significant in TB high-burden countries were examined. RESULTS: The 517 CXRs included were deemed abnormal by the three CAD with a sensitivity of respectively 97% (95% CI 95–98), 94% (95% CI 91–95), and 87% (95% CI 84–90) for INSIGHT CXR, qXR, and DrAID. The CAD generally detected abnormalities in patients with critical diagnoses such as lung cancer or heart failure. Performance for detecting other abnormalities was variable. CONCLUSION: This study showed that the three CAD tools identified CXRs as abnormal when diseases other than TB were present. Our findings alleviate ethical concerns of missing abnormalities other than TB when using commercially available CAD for TB screening and show their potential broader applicability.https://www.ingentaconnect.com/contentone/iuatld/ijtldo/2024/00000001/00000012/art00002tuberculosis screeningnon-communicable diseasesartificial intelligencechest x-raycad |
| spellingShingle | J. Sebastian I.D. Olaru A. Giannakis M. Arentz S.V. Kik M. Ruhwald S. Linsen G. Günther P. Wolf F.J. Herth T. Weber C.M. Denkinger Detection of other pathologies when utilising computer-assisted digital solutions for TB screening IJTLD Open tuberculosis screening non-communicable diseases artificial intelligence chest x-ray cad |
| title | Detection of other pathologies when utilising computer-assisted digital solutions for TB screening |
| title_full | Detection of other pathologies when utilising computer-assisted digital solutions for TB screening |
| title_fullStr | Detection of other pathologies when utilising computer-assisted digital solutions for TB screening |
| title_full_unstemmed | Detection of other pathologies when utilising computer-assisted digital solutions for TB screening |
| title_short | Detection of other pathologies when utilising computer-assisted digital solutions for TB screening |
| title_sort | detection of other pathologies when utilising computer assisted digital solutions for tb screening |
| topic | tuberculosis screening non-communicable diseases artificial intelligence chest x-ray cad |
| url | https://www.ingentaconnect.com/contentone/iuatld/ijtldo/2024/00000001/00000012/art00002 |
| work_keys_str_mv | AT jsebastian detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT idolaru detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT agiannakis detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT marentz detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT svkik detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT mruhwald detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT slinsen detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT ggunther detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT pwolf detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT fjherth detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT tweber detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening AT cmdenkinger detectionofotherpathologieswhenutilisingcomputerassisteddigitalsolutionsfortbscreening |