A deep learning model effectively identified acromegaly using hand images, preserving patient privacy by excluding facial and fingerprint features.
2
The model achieved 89% sensitivity and 91% specificity, outperforming board-certified endocrinologists in acromegaly detection.
3
The study involved 716 patients, including 317 with acromegaly, recruited from 15 Japanese pituitary centers.
4
Subgroup analyses showed consistent model performance across age, remission status, and sex, indicating robust diagnostic capability.
5
Limitations include potential generalizability issues due to the study's focus on Japanese patients and incomplete biochemical confirmation.
AACE Endocrine AI
is published by Conexiant under a license arrangement with the American Association of Clinical Endocrinology, Inc. (AACE®). The ideas and opinions expressed in
AACE Endocrine AI do not necessarily reflect those of Conexiant or AACE. For more information, see Policies.