Detection of Acromegaly From Facial Images Using Machine Learning: A Comparison With Clinical Experts
A deep learning model pretrained on facial features (FaRL) can detect acromegaly from standard photographs with accuracy comparable to that of expert endocrinologists. This supports the feasibility of face analysis as a screening tool for acromegaly.
Authors
- Maria Petersson (ORCID: https://orcid.org/0000-0002-2970-8593)
- Anna‐Karin Åkerman (ORCID: https://orcid.org/0000-0003-3190-7695)
- Pia Burman (ORCID: https://orcid.org/0000-0002-4844-8336)
- Charlotte Höybye (ORCID: https://orcid.org/0000-0002-3980-1927)
- Óskar Ragnarsson (ORCID: https://orcid.org/0000-0003-0204-9492)
- Erik Ylipää (ORCID: https://orcid.org/0000-0001-5027-1552)
- Bertil Ekman (ORCID: https://orcid.org/0000-0001-8732-7361)
- Sophie Bensing (ORCID: https://orcid.org/0000-0002-9193-2860)
- Henrik Borg (ORCID: https://orcid.org/0000-0002-4289-5557)
- Tommy Olsson (ORCID: https://orcid.org/0000-0001-7768-1076)
- Per Dahlqvist (ORCID: https://orcid.org/0000-0002-6471-9503)
- Konstantina Vouzouneraki (ORCID: https://orcid.org/0000-0002-9501-6763)
- Katarina Berinder (ORCID: https://orcid.org/0000-0003-2488-0375)
- Britt Edén Engström (ORCID: https://orcid.org/0000-0003-4514-5193)
- Jonas Robèrt (ORCID: https://orcid.org/0009-0005-3598-4155)
Institutions
- Linköping University (SE)
- Karolinska University Hospital (SE)
- Uppsala University Hospital (SE)
- Lund University (SE)
- Örebro University (SE)
- Sahlgrenska University Hospital (SE)
- Karolinska Institutet (SE)
- Wallenberg Wood Science Center (SE)
- Skåne University Hospital (SE)
- University of Gothenburg (SE)
- Umeå University (SE)
Publication Details
- Journal
- Journal of the Endocrine Society
- Published
- 2025-12-09
- DOI
- https://doi.org/10.1210/jendso/bvaf203
- Primary Topic
- Pituitary Gland Disorders and Treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00