Multi‐task <scp>AI</scp> models in dermatology: Overcoming critical clinical translation challenges for enhanced skin lesion diagnosis
Our HOT-AI model attempts to address common challenges in existing models by combining three tasks in one model to increase accuracy and clinical utility. By providing a more nuanced prediction, and alert for OOD, the model output provides greater explainability of the AI decision process. Prospective clinical testing is required to measure how this additional output impacts user trust, and how the model performs in a real-world setting.
Authors
- Maithili Sashindranath (ORCID: https://orcid.org/0000-0002-9712-4784)
- Clare Primiero (ORCID: https://orcid.org/0000-0002-2944-0013)
- Brigid Betz‐Stablein (ORCID: https://orcid.org/0000-0003-3876-501X)
- H. Peter Soyer (ORCID: https://orcid.org/0000-0002-4770-561X)
- Zongyuan Ge (ORCID: https://orcid.org/0000-0002-5880-8673)
- Martin Haskett (ORCID: https://orcid.org/0000-0002-3357-5826)
- Toan D. Nguyen (ORCID: https://orcid.org/0000-0001-6192-8601)
- Deval Mehta (ORCID: https://orcid.org/0000-0002-6907-7589)
- Victoria Mar (ORCID: https://orcid.org/0000-0001-9423-3435)
- Adrian Bowling (ORCID: https://orcid.org/0000-0001-6324-9610)
- C. Paul Bonnington (ORCID: https://orcid.org/0000-0002-9171-6949)
- Yaniv Gal
Institutions
- The University of Queensland (AU)
- Melanoma Institute Australia (AU)
- Australian Centre for HIV and Hepatitis Virology Research (AU)
- Monash University (AU)
Publication Details
- Journal
- Journal of the European Academy of Dermatology and Venereology
- Published
- 2025-02-07
- DOI
- https://doi.org/10.1111/jdv.20551
- Citations
- 2
- Primary Topic
- Cutaneous Melanoma Detection and Management
- Type
- article
- Field-Weighted Citation Impact
- 2.09
Funders
- Medical Research Council
- National Health and Medical Research Council