Hand Photos May Flag Rare Hormone Disorder

Deep learning model identifies disease using dorsal hand images

March 23, 2026
3 min

The Pathologist

Share

A deep learning model developed by researchers analyzes hand images to detect acromegaly, which is often misdiagnosed due to gradual physical changes. The multicenter observational study, published in The Journal of Clinical Endocrinology & Metabolism, included 716 participants and demonstrated high sensitivity (0.89) and specificity (0.91) in identifying the disorder. By leveraging convolutional neural networks, the study suggests this AI tool could enhance early diagnosis and support clinical evaluation, especially in subtle cases of acromegaly.