Benchmarking the most popular XAI used for explaining clinical predictive models: Untrustworthy but could be useful
Explanations are not trustworthy to guide clinical interventions, though they may offer useful insights and help model troubleshooting. Clinician-informed XAI development and presentation, clear disclaimers on limitations, and critical clinical judgment can promote informed decisions and prevent over-reliance.
Authors
- Aida Brankovic (ORCID: https://orcid.org/0000-0001-7978-575X)
- Jessica Sharmin Rahman (ORCID: https://orcid.org/0000-0001-9930-241X)
- David Cook (ORCID: https://orcid.org/0000-0001-9761-0179)
- Sankalp Khanna (ORCID: https://orcid.org/0000-0001-6125-8871)
- W. Huang (ORCID: https://orcid.org/0009-0000-0443-7902)
Institutions
- Commonwealth Scientific and Industrial Research Organisation (AU)
- The University of Queensland (AU)
- Princess Alexandra Hospital (AU)
- Australian e-Health Research Centre (AU)
Publication Details
- Journal
- Health Informatics Journal
- Published
- 2024-10-01
- DOI
- https://doi.org/10.1177/14604582241304730
- Citations
- 8
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 2.78